U.S. patent application number 17/549713 was filed with the patent office on 2022-04-07 for technolgies for millimeter wave rack interconnects.
The applicant listed for this patent is Intel Corporation. Invention is credited to Matthew J. ADILETTA, Michael T. CROCKER, Paul H. DORMITZER, Aaron GORIUS, Mark A. SCHMISSEUR, Myles WILDE.
Application Number | 20220107741 17/549713 |
Document ID | / |
Family ID | 1000006025925 |
Filed Date | 2022-04-07 |
![](/patent/app/20220107741/US20220107741A1-20220407-D00000.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00001.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00002.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00003.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00004.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00005.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00006.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00007.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00008.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00009.png)
![](/patent/app/20220107741/US20220107741A1-20220407-D00010.png)
View All Diagrams
United States Patent
Application |
20220107741 |
Kind Code |
A1 |
ADILETTA; Matthew J. ; et
al. |
April 7, 2022 |
TECHNOLGIES FOR MILLIMETER WAVE RACK INTERCONNECTS
Abstract
Racks and rack pods to support a plurality of sleds are
disclosed herein. Switches for use in the rack pods are also
disclosed herein. A rack comprises a plurality of sleds and a
plurality of electromagnetic waveguides. The plurality of sleds are
vertically spaced from one another. The plurality of
electromagnetic waveguides communicate data signals between the
plurality of sleds.
Inventors: |
ADILETTA; Matthew J.;
(Bolton, MA) ; WILDE; Myles; (Charlestown, MA)
; GORIUS; Aaron; (Upton, MA) ; CROCKER; Michael
T.; (Portland, OR) ; DORMITZER; Paul H.;
(Acton, MA) ; SCHMISSEUR; Mark A.; (Phoenix,
AZ) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Intel Corporation |
Santa Clara |
CA |
US |
|
|
Family ID: |
1000006025925 |
Appl. No.: |
17/549713 |
Filed: |
December 13, 2021 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
16346341 |
Apr 30, 2019 |
11200104 |
|
|
PCT/US17/63759 |
Nov 29, 2017 |
|
|
|
17549713 |
|
|
|
|
62584401 |
Nov 10, 2017 |
|
|
|
62427268 |
Nov 29, 2016 |
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06F 21/76 20130101;
G06F 3/0608 20130101; H04L 41/0853 20130101; H04L 12/2881 20130101;
H01R 13/453 20130101; H03K 19/1731 20130101; G06T 9/005 20130101;
G06F 3/0613 20130101; G06F 3/0653 20130101; H04L 41/142 20130101;
G06F 9/4843 20130101; G06F 9/5038 20130101; G06F 7/06 20130101;
H04L 61/2007 20130101; H04L 47/20 20130101; G06F 3/0641 20130101;
H04L 41/046 20130101; H01R 13/4538 20130101; G06F 3/0611 20130101;
G06F 12/0692 20130101; G06F 9/4401 20130101; H04L 43/0894 20130101;
G06F 11/0709 20130101; H04L 41/12 20130101; H05K 7/1492 20130101;
H04L 47/78 20130101; H05K 7/1447 20130101; H03M 7/6029 20130101;
G06T 1/60 20130101; G06F 8/654 20180201; G06F 3/0647 20130101; H04L
67/1014 20130101; G06F 12/0284 20130101; H04L 49/104 20130101; G06F
21/6218 20130101; G06F 2221/2107 20130101; G06F 11/3006 20130101;
G06F 11/3055 20130101; H03M 7/3084 20130101; G06T 1/20 20130101;
H04L 43/04 20130101; G06F 9/5005 20130101; G06F 2212/401 20130101;
G06F 9/45533 20130101; H03M 7/40 20130101; H04L 67/327 20130101;
G06F 3/0604 20130101; G06F 15/80 20130101; G06F 11/3034 20130101;
G06F 13/1652 20130101; G06F 11/079 20130101; G06F 21/73 20130101;
H03M 7/60 20130101; H03M 7/6017 20130101; H05K 7/1487 20130101;
H04L 43/06 20130101; G06F 3/065 20130101; H05K 7/1491 20130101;
H03M 7/6011 20130101; G06F 12/023 20130101; H01R 13/631 20130101;
H03M 7/42 20130101; G06F 9/544 20130101; G06F 8/65 20130101; G06F
9/3851 20130101; G06F 8/656 20180201; H01R 13/4536 20130101; G06F
8/658 20180201; H05K 7/1452 20130101; H04L 67/36 20130101; G06F
9/4881 20130101; G06F 11/3409 20130101; H04Q 11/0005 20130101; H04L
41/044 20130101; H04L 41/0816 20130101; G06F 9/5083 20130101; G06F
16/285 20190101; H04L 9/0822 20130101; G06F 11/3079 20130101; H04L
63/1425 20130101; G06F 3/0617 20130101; G06F 9/505 20130101; G06F
21/57 20130101; H04L 12/4633 20130101; H04L 41/0896 20130101; H04L
47/2441 20130101; G06F 9/3891 20130101; H04L 67/10 20130101; G06F
2212/402 20130101; G06F 11/0751 20130101; H04L 43/08 20130101; G06F
16/1744 20190101; G06F 3/067 20130101; G06F 11/1453 20130101; G06F
9/5044 20130101 |
International
Class: |
G06F 3/06 20060101
G06F003/06; G06F 16/174 20060101 G06F016/174; G06F 21/57 20060101
G06F021/57; G06F 21/73 20060101 G06F021/73; G06F 8/65 20060101
G06F008/65; H04L 12/24 20060101 H04L012/24; H04L 29/08 20060101
H04L029/08; G06F 11/30 20060101 G06F011/30; G06F 9/50 20060101
G06F009/50; H01R 13/453 20060101 H01R013/453; G06F 9/48 20060101
G06F009/48; G06F 9/455 20060101 G06F009/455; H05K 7/14 20060101
H05K007/14; H03M 7/30 20060101 H03M007/30; H03M 7/40 20060101
H03M007/40; H04L 12/26 20060101 H04L012/26; H04L 12/813 20060101
H04L012/813; H04L 12/851 20060101 H04L012/851; G06F 11/07 20060101
G06F011/07; G06F 11/34 20060101 G06F011/34; G06F 7/06 20060101
G06F007/06; G06T 9/00 20060101 G06T009/00; H03M 7/42 20060101
H03M007/42; H04L 12/28 20060101 H04L012/28; H04L 12/46 20060101
H04L012/46; H04L 29/12 20060101 H04L029/12; G06F 13/16 20060101
G06F013/16; G06F 21/62 20060101 G06F021/62; G06F 21/76 20060101
G06F021/76; H03K 19/173 20060101 H03K019/173; H04L 9/08 20060101
H04L009/08; H04L 12/933 20060101 H04L012/933; G06F 9/38 20060101
G06F009/38; G06F 12/02 20060101 G06F012/02; G06F 12/06 20060101
G06F012/06; G06T 1/20 20060101 G06T001/20; G06T 1/60 20060101
G06T001/60; G06F 9/54 20060101 G06F009/54; G06F 8/656 20060101
G06F008/656; G06F 8/658 20060101 G06F008/658; G06F 8/654 20060101
G06F008/654; G06F 9/4401 20060101 G06F009/4401; H01R 13/631
20060101 H01R013/631 |
Foreign Application Data
Date |
Code |
Application Number |
Aug 30, 2017 |
IN |
201741030632 |
Claims
1. A system comprising: a server; a switch; and a wave guide to
provide communications between the server and the switch.
2. The system of claim 1, wherein the waveguide is to communicate
millimeter wave data signals between the server and the switch at
about 50 to 100 gigabits per second.
3. The system of claim 1, wherein millimeter wave data signals have
a carrier frequency range of about 60 to 120 gigahertz.
4. The system of claim 1, wherein the server comprises a set of
compute circuit boards.
5. The system of claim 1, wherein the server comprises a set of
memory circuit boards.
6. The system of claim 1, wherein the switch comprises at least one
circuit board.
7. The system of claim 1, comprising: a circuit board coupled to
the switch.
8. The system of claim 7, wherein the circuit board comprises at
least one processor and at least one memory device.
9. A rack comprising: a plurality of servers; a switch; and a wave
guide to provide communications between the plurality of servers
and the switch.
10. The rack of claim 9, wherein the waveguide is to communicate
millimeter wave data signals between the server and the switch at
about 50 to 100 gigabits per second.
11. The rack of claim 9, wherein millimeter wave data signals have
a carrier frequency range of about 60 to 120 gigahertz.
12. The rack of claim 9, wherein at least one server of the
plurality of servers comprises a set of compute circuit boards.
13. The rack of claim 9, wherein at least one server of the
plurality of servers comprises a set of memory circuit boards.
14. The rack of claim 9, wherein the switch comprises at least one
circuit board.
15. The rack of claim 9, comprising: a circuit board coupled to the
switch.
16. The rack of claim 15, wherein the circuit board comprises at
least one processor and at least one memory device.
Description
CROSS-REFERENCE TO RELATED APPLICATION
[0001] The present application is a national stage entry under 35
USC .sctn. 371(b) of International Application No.
PCT/US2017/063759, filed Nov. 29, 2017 and claims the benefit of
U.S. Provisional Patent Application No. 62/427,268, filed Nov. 29,
2016, Indian Provisional Patent Application No. 201741030632, filed
Aug. 30, 2017, and U.S. Provisional Patent Application No.
62/584,401, filed Nov. 10, 2017.
BACKGROUND
[0002] Typical enterprise-level data centers can include several to
hundreds of racks or cabinets, with each rack/cabinet housing
multiple servers. Each of the various servers of a data center may
be communicatively connectable to each other via one or more local
networking switches, routers, and/or other interconnecting devices,
cables, and/or interfaces. The number of racks and servers of a
particular data center, as well as the complexity of the design of
the data center, may depend on the intended use of the data center,
as well as the quality of service the data center is intended to
provide.
[0003] Traditional rack systems are self-contained physical support
structures that include a number of pre-defined server spaces. A
corresponding server may be mounted in each pre-defined server
space. When mounted in the server spaces of a single rack, the
servers may be spaced such that communicatively coupling the
servers to one another by conventional electrical cables or printed
circuit boards is impractical. Additionally, when servers are
mounted in the server spaces of multiple racks that are spaced from
one another, communicatively coupling the servers to one another by
such devices may not be feasible.
BRIEF DESCRIPTION OF THE DRAWINGS
[0004] The concepts described herein are illustrated by way of
example and not by way of limitation in the accompanying figures.
For simplicity and clarity of illustration, elements illustrated in
the figures are not necessarily drawn to scale. Where considered
appropriate, reference labels have been repeated among the figures
to indicate corresponding or analogous elements.
[0005] FIG. 1 is a diagram of a conceptual overview of a data
center in which one or more techniques described herein may be
implemented according to various embodiments;
[0006] FIG. 2 is a diagram of an example embodiment of a logical
configuration of a rack of the data center of FIG. 1;
[0007] FIG. 3 is a diagram of an example embodiment of another data
center in which one or more techniques described herein may be
implemented according to various embodiments;
[0008] FIG. 4 is a diagram of another example embodiment of a data
center in which one or more techniques described herein may be
implemented according to various embodiments;
[0009] FIG. 5 is a diagram of a connectivity scheme representative
of link-layer connectivity that may be established among various
sleds of the data centers of FIGS. 1, 3, and 4;
[0010] FIG. 6 is a diagram of a rack architecture that may be
representative of an architecture of any particular one of the
racks depicted in FIGS. 1-4 according to some embodiments;
[0011] FIG. 7 is a diagram of an example embodiment of a sled that
may be used with the rack architecture of FIG. 6;
[0012] FIG. 8 is a diagram of an example embodiment of a rack
architecture to provide support for sleds featuring expansion
capabilities;
[0013] FIG. 9 is a diagram of an example embodiment of a rack
implemented according to the rack architecture of FIG. 8;
[0014] FIG. 10 is a diagram of an example embodiment of a sled
designed for use in conjunction with the rack of FIG. 9;
[0015] FIG. 11 is a diagram of an example embodiment of a data
center in which one or more techniques described herein may be
implemented according to various embodiments;
[0016] FIG. 12 is a partial perspective view of at least one
embodiment of a rack of the data center of FIGS. 1, 3, and 4 with
multiple compute sleds each communicatively coupled to a memory
sled by an electromagnetic waveguide;
[0017] FIG. 13 is a front elevation view of at least one embodiment
of a rack pod of the data center of FIGS. 1, 3, and 4 with a switch
arranged between a first rack and a second rack;
[0018] FIG. 14 is a perspective view of the switch shown in FIG.
13;
[0019] FIG. 15 is a perspective view of a connector of an
electromagnetic waveguide that is configured to mate with a
corresponding waveguide connector of the switch of FIG. 14;
[0020] FIG. 16 is a front elevation view of a waveguide connector
of the switch of FIG. 14;
[0021] FIG. 17 is a diagram of a connectivity scheme that may be
established among various sleds of the data centers of FIGS. 1, 3,
4, 12, and 13;
[0022] FIG. 18 is a front perspective view of a pod switch that may
be included in the data centers of FIGS. 1, 3, 4, 12, and 13;
[0023] FIG. 19 is a rear perspective view of the pod switch of FIG.
18; and
[0024] FIG. 20 is a partial elevation view of the pod switch of
FIG. 19 with a first set of line cards communicatively coupled to a
second set of line cards by electromagnetic waveguides.
DETAILED DESCRIPTION OF THE DRAWINGS
[0025] While the concepts of the present disclosure are susceptible
to various modifications and alternative forms, specific
embodiments thereof have been shown by way of example in the
drawings and will be described herein in detail. It should be
understood, however, that there is no intent to limit the concepts
of the present disclosure to the particular forms disclosed, but on
the contrary, the intention is to cover all modifications,
equivalents, and alternatives consistent with the present
disclosure and the appended claims.
[0026] References in the specification to "one embodiment," "an
embodiment," "an illustrative embodiment," etc., indicate that the
embodiment described may include a particular feature, structure,
or characteristic, but every embodiment may or may not necessarily
include that particular feature, structure, or characteristic.
Moreover, such phrases are not necessarily referring to the same
embodiment. Further, when a particular feature, structure, or
characteristic is described in connection with an embodiment, it is
submitted that it is within the knowledge of one skilled in the art
to effect such feature, structure, or characteristic in connection
with other embodiments whether or not explicitly described.
Additionally, it should be appreciated that items included in a
list in the form of "at least one A, B, and C" can mean (A); (B);
(C); (A and B); (A and C); (B and C); or (A, B, and C). Similarly,
items listed in the form of "at least one of A, B, or C" can mean
(A); (B); (C); (A and B); (A and C); (B and C); or (A, B, and
C).
[0027] The disclosed embodiments may be implemented, in some cases,
in hardware, firmware, software, or any combination thereof. The
disclosed embodiments may also be implemented as instructions
carried by or stored on a transitory or non-transitory
machine-readable (e.g., computer-readable) storage medium, which
may be read and executed by one or more processors. A
machine-readable storage medium may be embodied as any storage
device, mechanism, or other physical structure for storing or
transmitting information in a form readable by a machine (e.g., a
volatile or non-volatile memory, a media disc, or other media
device).
[0028] In the drawings, some structural or method features may be
shown in specific arrangements and/or orderings. However, it should
be appreciated that such specific arrangements and/or orderings may
not be required. Rather, in some embodiments, such features may be
arranged in a different manner and/or order than shown in the
illustrative figures. Additionally, the inclusion of a structural
or method feature in a particular figure is not meant to imply that
such feature is required in all embodiments and, in some
embodiments, may not be included or may be combined with other
features.
[0029] FIG. 1 illustrates a conceptual overview of a data center
100 that may generally be representative of a data center or other
type of computing network in/for which one or more techniques
described herein may be implemented according to various
embodiments. As shown in FIG. 1, data center 100 may generally
contain a plurality of racks, each of which may house computing
equipment comprising a respective set of physical resources. In the
particular non-limiting example depicted in FIG. 1, data center 100
contains four racks 102A to 102D, which house computing equipment
comprising respective sets of physical resources (PCRs) 105A to
105D. According to this example, a collective set of physical
resources 106 of data center 100 includes the various sets of
physical resources 105A to 105D that are distributed among racks
102A to 102D. Physical resources 106 may include resources of
multiple types, such as--for example--processors, co-processors,
accelerators, field programmable gate arrays (FPGAs), memory, and
storage. The embodiments are not limited to these examples.
[0030] The illustrative data center 100 differs from typical data
centers in many ways. For example, in the illustrative embodiment,
the circuit boards ("sleds") on which components such as CPUs,
memory, and other components are placed are designed for increased
thermal performance. In particular, in the illustrative embodiment,
the sleds are shallower than typical boards. In other words, the
sleds are shorter from the front to the back, where cooling fans
are located. This decreases the length of the path that air must to
travel across the components on the board. Further, the components
on the sled are spaced further apart than in typical circuit
boards, and the components are arranged to reduce or eliminate
shadowing (i.e., one component in the air flow path of another
component). In the illustrative embodiment, processing components
such as the processors are located on a top side of a sled while
near memory, such as DIMMs, are located on a bottom side of the
sled. As a result of the enhanced airflow provided by this design,
the components may operate at higher frequencies and power levels
than in typical systems, thereby increasing performance.
Furthermore, the sleds are configured to blindly mate with power
and data communication cables in each rack 102A, 102B, 102C, 102D,
enhancing their ability to be quickly removed, upgraded,
reinstalled, and/or replaced. Similarly, individual components
located on the sleds, such as processors, accelerators, memory, and
data storage drives, are configured to be easily upgraded due to
their increased spacing from each other. In the illustrative
embodiment, the components additionally include hardware
attestation features to prove their authenticity.
[0031] Furthermore, in the illustrative embodiment, the data center
100 utilizes a single network architecture ("fabric") that supports
multiple other network architectures including Ethernet and
Omni-Path. The sleds, in the illustrative embodiment, are coupled
to switches via optical fibers, which provide higher bandwidth and
lower latency than typical twisted pair cabling (e.g., Category 5,
Category Se, Category 6, etc.). Due to the high bandwidth, low
latency interconnections and network architecture, the data center
100 may, in use, pool resources, such as memory, accelerators
(e.g., graphics accelerators, FPGAs, ASICs, etc.), and data storage
drives that are physically disaggregated, and provide them to
compute resources (e.g., processors) on an as needed basis,
enabling the compute resources to access the pooled resources as if
they were local. The illustrative data center 100 additionally
receives utilization information for the various resources,
predicts resource utilization for different types of workloads
based on past resource utilization, and dynamically reallocates the
resources based on this information.
[0032] The racks 102A, 102B, 102C, 102D of the data center 100 may
include physical design features that facilitate the automation of
a variety of types of maintenance tasks. For example, data center
100 may be implemented using racks that are designed to be
robotically-accessed, and to accept and house
robotically-manipulatable resource sleds. Furthermore, in the
illustrative embodiment, the racks 102A, 102B, 102C, 102D include
integrated power sources that receive a greater voltage than is
typical for power sources. The increased voltage enables the power
sources to provide additional power to the components on each sled,
enabling the components to operate at higher than typical
frequencies.
[0033] FIG. 2 illustrates an exemplary logical configuration of a
rack 202 of the data center 100. As shown in FIG. 2, rack 202 may
generally house a plurality of sleds, each of which may comprise a
respective set of physical resources. In the particular
non-limiting example depicted in FIG. 2, rack 202 houses sleds
204-1 to 204-4 comprising respective sets of physical resources
205-1 to 205-4, each of which constitutes a portion of the
collective set of physical resources 206 comprised in rack 202.
With respect to FIG. 1, if rack 202 is representative of--for
example--rack 102A, then physical resources 206 may correspond to
the physical resources 105A comprised in rack 102A. In the context
of this example, physical resources 105A may thus be made up of the
respective sets of physical resources, including physical storage
resources 205-1, physical accelerator resources 205-2, physical
memory resources 205-3, and physical compute resources 205-5
comprised in the sleds 204-1 to 204-4 of rack 202. The embodiments
are not limited to this example. Each sled may contain a pool of
each of the various types of physical resources (e.g., compute,
memory, accelerator, storage). By having robotically accessible and
robotically manipulatable sleds comprising disaggregated resources,
each type of resource can be upgraded independently of each other
and at their own optimized refresh rate.
[0034] FIG. 3 illustrates an example of a data center 300 that may
generally be representative of one in/for which one or more
techniques described herein may be implemented according to various
embodiments. In the particular non-limiting example depicted in
FIG. 3, data center 300 comprises racks 302-1 to 302-32. In various
embodiments, the racks of data center 300 may be arranged in such
fashion as to define and/or accommodate various access pathways.
For example, as shown in FIG. 3, the racks of data center 300 may
be arranged in such fashion as to define and/or accommodate access
pathways 311A, 311B, 311C, and 311D. In some embodiments, the
presence of such access pathways may generally enable automated
maintenance equipment, such as robotic maintenance equipment, to
physically access the computing equipment housed in the various
racks of data center 300 and perform automated maintenance tasks
(e.g., replace a failed sled, upgrade a sled). In various
embodiments, the dimensions of access pathways 311A, 311B, 311C,
and 311D, the dimensions of racks 302-1 to 302-32, and/or one or
more other aspects of the physical layout of data center 300 may be
selected to facilitate such automated operations. The embodiments
are not limited in this context.
[0035] FIG. 4 illustrates an example of a data center 400 that may
generally be representative of one in/for which one or more
techniques described herein may be implemented according to various
embodiments. As shown in FIG. 4, data center 400 may feature an
optical fabric 412. Optical fabric 412 may generally comprise a
combination of optical signaling media (such as optical cabling)
and optical switching infrastructure via which any particular sled
in data center 400 can send signals to (and receive signals from)
each of the other sleds in data center 400. The signaling
connectivity that optical fabric 412 provides to any given sled may
include connectivity both to other sleds in a same rack and sleds
in other racks. In the particular non-limiting example depicted in
FIG. 4, data center 400 includes four racks 402A to 402D. Racks
402A to 402D house respective pairs of sleds 404A-1 and 404A-2,
404B-1 and 404B-2, 404C-1 and 404C-2, and 404D-1 and 404D-2. Thus,
in this example, data center 400 comprises a total of eight sleds.
Via optical fabric 412, each such sled may possess signaling
connectivity with each of the seven other sleds in data center 400.
For example, via optical fabric 412, sled 404A-1 in rack 402A may
possess signaling connectivity with sled 404A-2 in rack 402A, as
well as the six other sleds 404B-1, 404B-2, 404C-1, 404C-2, 404D-1,
and 404D-2 that are distributed among the other racks 402B, 402C,
and 402D of data center 400. The embodiments are not limited to
this example.
[0036] FIG. 5 illustrates an overview of a connectivity scheme 500
that may generally be representative of link-layer connectivity
that may be established in some embodiments among the various sleds
of a data center, such as any of example data centers 100, 300, and
400 of FIGS. 1, 3, and 4. Connectivity scheme 500 may be
implemented using an optical fabric that features a dual-mode
optical switching infrastructure 514. Dual-mode optical switching
infrastructure 514 may generally comprise a switching
infrastructure that is capable of receiving communications
according to multiple link-layer protocols via a same unified set
of optical signaling media, and properly switching such
communications. In various embodiments, dual-mode optical switching
infrastructure 514 may be implemented using one or more dual-mode
optical switches 515. In various embodiments, dual-mode optical
switches 515 may generally comprise high-radix switches. In some
embodiments, dual-mode optical switches 515 may comprise multi-ply
switches, such as four-ply switches. In various embodiments,
dual-mode optical switches 515 may feature integrated silicon
photonics that enable them to switch communications with
significantly reduced latency in comparison to conventional
switching devices. In some embodiments, dual-mode optical switches
515 may constitute leaf switches 530 in a leaf-spine architecture
additionally including one or more dual-mode optical spine switches
520.
[0037] In various embodiments, dual-mode optical switches may be
capable of receiving both Ethernet protocol communications carrying
Internet Protocol (IP packets) and communications according to a
second, high-performance computing (HPC) link-layer protocol (e.g.,
Intel's Omni-Path Architecture's, Infiniband) via optical signaling
media of an optical fabric. As reflected in FIG. 5, with respect to
any particular pair of sleds 504A and 504B possessing optical
signaling connectivity to the optical fabric, connectivity scheme
500 may thus provide support for link-layer connectivity via both
Ethernet links and HPC links. Thus, both Ethernet and HPC
communications can be supported by a single high-bandwidth,
low-latency switch fabric. The embodiments are not limited to this
example.
[0038] FIG. 6 illustrates a general overview of a rack architecture
600 that may be representative of an architecture of any particular
one of the racks depicted in FIGS. 1 to 4 according to some
embodiments. As reflected in FIG. 6, rack architecture 600 may
generally feature a plurality of sled spaces into which sleds may
be inserted, each of which may be robotically-accessible via a rack
access region 601. In the particular non-limiting example depicted
in FIG. 6, rack architecture 600 features five sled spaces 603-1 to
603-5. Sled spaces 603-1 to 603-5 feature respective multi-purpose
connector modules (MPCMs) 616-1 to 616-5.
[0039] FIG. 7 illustrates an example of a sled 704 that may be
representative of a sled of such a type. As shown in FIG. 7, sled
704 may comprise a set of physical resources 705, as well as an
MPCM 716 designed to couple with a counterpart MPCM when sled 704
is inserted into a sled space such as any of sled spaces 603-1 to
603-5 of FIG. 6. Sled 704 may also feature an expansion connector
717. Expansion connector 717 may generally comprise a socket, slot,
or other type of connection element that is capable of accepting
one or more types of expansion modules, such as an expansion sled
718. By coupling with a counterpart connector on expansion sled
718, expansion connector 717 may provide physical resources 705
with access to supplemental computing resources 705B residing on
expansion sled 718. The embodiments are not limited in this
context.
[0040] FIG. 8 illustrates an example of a rack architecture 800
that may be representative of a rack architecture that may be
implemented in order to provide support for sleds featuring
expansion capabilities, such as sled 704 of FIG. 7. In the
particular non-limiting example depicted in FIG. 8, rack
architecture 800 includes seven sled spaces 803-1 to 803-7, which
feature respective MPCMs 816-1 to 816-7. Sled spaces 803-1 to 803-7
include respective primary regions 803-1A to 803-7A and respective
expansion regions 803-1B to 803-7B. With respect to each such sled
space, when the corresponding MPCM is coupled with a counterpart
MPCM of an inserted sled, the primary region may generally
constitute a region of the sled space that physically accommodates
the inserted sled. The expansion region may generally constitute a
region of the sled space that can physically accommodate an
expansion module, such as expansion sled 718 of FIG. 7, in the
event that the inserted sled is configured with such a module.
[0041] FIG. 9 illustrates an example of a rack 902 that may be
representative of a rack implemented according to rack architecture
800 of FIG. 8 according to some embodiments. In the particular
non-limiting example depicted in FIG. 9, rack 902 features seven
sled spaces 903-1 to 903-7, which include respective primary
regions 903-1A to 903-7A and respective expansion regions 903-1B to
903-7B. In various embodiments, temperature control in rack 902 may
be implemented using an air cooling system. For example, as
reflected in FIG. 9, rack 902 may feature a plurality of fans 919
that are generally arranged to provide air cooling within the
various sled spaces 903-1 to 903-7. In some embodiments, the height
of the sled space is greater than the conventional "1 U" server
height. In such embodiments, fans 919 may generally comprise
relatively slow, large diameter cooling fans as compared to fans
used in conventional rack configurations. Running larger diameter
cooling fans at lower speeds may increase fan lifetime relative to
smaller diameter cooling fans running at higher speeds while still
providing the same amount of cooling. The sleds are physically
shallower than conventional rack dimensions. Further, components
are arranged on each sled to reduce thermal shadowing (i.e., not
arranged serially in the direction of air flow). As a result, the
wider, shallower sleds allow for an increase in device performance
because the devices can be operated at a higher thermal envelope
(e.g., 250 W) due to improved cooling (i.e., no thermal shadowing,
more space between devices, more room for larger heat sinks,
etc.).
[0042] MPCMs 916-1 to 916-7 may be configured to provide inserted
sleds with access to power sourced by respective power modules
920-1 to 920-7, each of which may draw power from an external power
source 921. In various embodiments, external power source 921 may
deliver alternating current (AC) power to rack 902, and power
modules 920-1 to 920-7 may be configured to convert such AC power
to direct current (DC) power to be sourced to inserted sleds. In
some embodiments, for example, power modules 920-1 to 920-7 may be
configured to convert 277-volt AC power into 12-volt DC power for
provision to inserted sleds via respective MPCMs 916-1 to 916-7.
The embodiments are not limited to this example.
[0043] MPCMs 916-1 to 916-7 may also be arranged to provide
inserted sleds with optical signaling connectivity to a dual-mode
optical switching infrastructure 914, which may be the same as--or
similar to--dual-mode optical switching infrastructure 514 of FIG.
5. In various embodiments, optical connectors contained in MPCMs
916-1 to 916-7 may be designed to couple with counterpart optical
connectors contained in MPCMs of inserted sleds to provide such
sleds with optical signaling connectivity to dual-mode optical
switching infrastructure 914 via respective lengths of optical
cabling 922-1 to 922-7. In some embodiments, each such length of
optical cabling may extend from its corresponding MPCM to an
optical interconnect loom 923 that is external to the sled spaces
of rack 902. In various embodiments, optical interconnect loom 923
may be arranged to pass through a support post or other type of
load-bearing element of rack 902. The embodiments are not limited
in this context. Because inserted sleds connect to an optical
switching infrastructure via MPCMs, the resources typically spent
in manually configuring the rack cabling to accommodate a newly
inserted sled can be saved.
[0044] FIG. 10 illustrates an example of a sled 1004 that may be
representative of a sled designed for use in conjunction with rack
902 of FIG. 9 according to some embodiments. Sled 1004 may feature
an MPCM 1016 that comprises an optical connector 1016A and a power
connector 1016B, and that is designed to couple with a counterpart
MPCM of a sled space in conjunction with insertion of MPCM 1016
into that sled space. Coupling MPCM 1016 with such a counterpart
MPCM may cause power connector 1016 to couple with a power
connector comprised in the counterpart MPCM. This may generally
enable physical resources 1005 of sled 1004 to source power from an
external source, via power connector 1016 and power transmission
media 1024 that conductively couples power connector 1016 to
physical resources 1005.
[0045] Sled 1004 may also include dual-mode optical network
interface circuitry 1026. Dual-mode optical network interface
circuitry 1026 may generally comprise circuitry that is capable of
communicating over optical signaling media according to each of
multiple link-layer protocols supported by dual-mode optical
switching infrastructure 914 of FIG. 9. In some embodiments,
dual-mode optical network interface circuitry 1026 may be capable
both of Ethernet protocol communications and of communications
according to a second, high-performance protocol. In various
embodiments, dual-mode optical network interface circuitry 1026 may
include one or more optical transceiver modules 1027, each of which
may be capable of transmitting and receiving optical signals over
each of one or more optical channels. The embodiments are not
limited in this context.
[0046] Coupling MPCM 1016 with a counterpart MPCM of a sled space
in a given rack may cause optical connector 1016A to couple with an
optical connector comprised in the counterpart MPCM. This may
generally establish optical connectivity between optical cabling of
the sled and dual-mode optical network interface circuitry 1026,
via each of a set of optical channels 1025. Dual-mode optical
network interface circuitry 1026 may communicate with the physical
resources 1005 of sled 1004 via electrical signaling media 1028. In
addition to the dimensions of the sleds and arrangement of
components on the sleds to provide improved cooling and enable
operation at a relatively higher thermal envelope (e.g., 250 W), as
described above with reference to FIG. 9, in some embodiments, a
sled may include one or more additional features to facilitate air
cooling, such as a heatpipe and/or heat sinks arranged to dissipate
heat generated by physical resources 1005. It is worthy of note
that although the example sled 1004 depicted in FIG. 10 does not
feature an expansion connector, any given sled that features the
design elements of sled 1004 may also feature an expansion
connector according to some embodiments. The embodiments are not
limited in this context.
[0047] FIG. 11 illustrates an example of a data center 1100 that
may generally be representative of one in/for which one or more
techniques described herein may be implemented according to various
embodiments. As reflected in FIG. 11, a physical infrastructure
management framework 1150A may be implemented to facilitate
management of a physical infrastructure 1100A of data center 1100.
In various embodiments, one function of physical infrastructure
management framework 1150A may be to manage automated maintenance
functions within data center 1100, such as the use of robotic
maintenance equipment to service computing equipment within
physical infrastructure 1100A. In some embodiments, physical
infrastructure 1100A may feature an advanced telemetry system that
performs telemetry reporting that is sufficiently robust to support
remote automated management of physical infrastructure 1100A. In
various embodiments, telemetry information provided by such an
advanced telemetry system may support features such as failure
prediction/prevention capabilities and capacity planning
capabilities. In some embodiments, physical infrastructure
management framework 1150A may also be configured to manage
authentication of physical infrastructure components using hardware
attestation techniques. For example, robots may verify the
authenticity of components before installation by analyzing
information collected from a radio frequency identification (RFID)
tag associated with each component to be installed. The embodiments
are not limited in this context.
[0048] As shown in FIG. 11, the physical infrastructure 1100A of
data center 1100 may comprise an optical fabric 1112, which may
include a dual-mode optical switching infrastructure 1114. Optical
fabric 1112 and dual-mode optical switching infrastructure 1114 may
be the same as--or similar to--optical fabric 412 of FIG. 4 and
dual-mode optical switching infrastructure 514 of FIG. 5,
respectively, and may provide high-bandwidth, low-latency,
multi-protocol connectivity among sleds of data center 1100. As
discussed above, with reference to FIG. 1, in various embodiments,
the availability of such connectivity may make it feasible to
disaggregate and dynamically pool resources such as accelerators,
memory, and storage. In some embodiments, for example, one or more
pooled accelerator sleds 1130 may be included among the physical
infrastructure 1100A of data center 1100, each of which may
comprise a pool of accelerator resources--such as co-processors
and/or FPGAs, for example--that is globally accessible to other
sleds via optical fabric 1112 and dual-mode optical switching
infrastructure 1114.
[0049] In another example, in various embodiments, one or more
pooled storage sleds 1132 may be included among the physical
infrastructure 1100A of data center 1100, each of which may
comprise a pool of storage resources that is globally accessible to
other sleds via optical fabric 1112 and dual-mode optical switching
infrastructure 1114. In some embodiments, such pooled storage sleds
1132 may comprise pools of solid-state storage devices such as
solid-state drives (SSDs). In various embodiments, one or more
high-performance processing sleds 1134 may be included among the
physical infrastructure 1100A of data center 1100. In some
embodiments, high-performance processing sleds 1134 may comprise
pools of high-performance processors, as well as cooling features
that enhance air cooling to yield a higher thermal envelope of up
to 250 W or more. In various embodiments, any given
high-performance processing sled 1134 may feature an expansion
connector 1117 that can accept a far memory expansion sled, such
that the far memory that is locally available to that
high-performance processing sled 1134 is disaggregated from the
processors and near memory comprised on that sled. In some
embodiments, such a high-performance processing sled 1134 may be
configured with far memory using an expansion sled that comprises
low-latency SSD storage. The optical infrastructure allows for
compute resources on one sled to utilize remote accelerator/FPGA,
memory, and/or SSD resources that are disaggregated on a sled
located on the same rack or any other rack in the data center. The
remote resources can be located one switch jump away or two-switch
jumps away in the spine-leaf network architecture described above
with reference to FIG. 5. The embodiments are not limited in this
context.
[0050] In various embodiments, one or more layers of abstraction
may be applied to the physical resources of physical infrastructure
1100A in order to define a virtual infrastructure, such as a
software-defined infrastructure 1100B. In some embodiments, virtual
computing resources 1136 of software-defined infrastructure 1100B
may be allocated to support the provision of cloud services 1140.
In various embodiments, particular sets of virtual computing
resources 1136 may be grouped for provision to cloud services 1140
in the form of SDI services 1138. Examples of cloud services 1140
may include--without limitation--software as a service (SaaS)
services 1142, platform as a service (PaaS) services 1144, and
infrastructure as a service (IaaS) services 1146.
[0051] In some embodiments, management of software-defined
infrastructure 1100B may be conducted using a virtual
infrastructure management framework 1150B. In various embodiments,
virtual infrastructure management framework 1150B may be designed
to implement workload fingerprinting techniques and/or
machine-learning techniques in conjunction with managing allocation
of virtual computing resources 1136 and/or SDI services 1138 to
cloud services 1140. In some embodiments, virtual infrastructure
management framework 1150B may use/consult telemetry data in
conjunction with performing such resource allocation. In various
embodiments, an application/service management framework 1150C may
be implemented in order to provide QoS management capabilities for
cloud services 1140. The embodiments are not limited in this
context.
[0052] Referring now to FIGS. 12-20, various embodiments of sleds,
racks, rows, pods, and switches may be connected with use of
electromagnetic waveguides configured to carry millimeter wave
signals at a carrier frequency of, e.g., 60-120 GHz. Use of
millimeter wave signals in electromagnetic waveguides may allow for
high-bandwidth signals (e.g., 50-100 gigabits per second per
waveguide) over relatively long distances (e.g., 10 feet) at a
relatively low cost. In particular, millimeter wave signals in
electromagnetic waveguides may be able to carry signals longer than
electrical cables at similar bitrates and may be implemented more
cheaply than an optical connection. For example, in a first
embodiment, electromagnetic waveguides 1206 carrying millimeter
wave signals may be used to connect each compute sled sled 1204C in
a rack 1202 to a central memory sled 1204M. In a second embodiment,
electromagnetic waveguides up to 10 feet long may be used to
connect a large number of sleds 1304 (such as 128 sleds 1304) to a
single switch coupling point 1352. In a third embodiment,
electromagnetic waveguides 2006 may provide backplane connections
to several line cards 1922 in a pod switch 1842.
[0053] Referring now to FIG. 12, in another embodiment, an
illustrative data center 1200 includes one or more racks 1202
configured to house or otherwise receive one or more sleds 1204 for
mounting therein. The data center 1200 may generally be
representative of any type of data center or other type of
computing network. Accordingly, the data center 1200 may be similar
to, embodied as, or otherwise form a part of, the data centers 100,
300, 400, 1100 described above. The rack 1202 may house computing
equipment comprising a set of physical resources, which may include
processors, co-processors, accelerators, field programmable gate
arrays (FPGAs), memory, and storage, for example. The rack 1202 may
therefore be similar to, embodied as, or otherwise form a part of,
the racks 102A-102D, 202, 302-1-302-32, 402A-402D, 902 described
above. Additionally, in some embodiments, the rack 1202 may
incorporate architecture similar to the aforementioned rack
architectures 600, 800. Each of the sleds 1204 may be embodied as a
circuit board on which components such as CPUs, memory, and/or
other components are placed. As such, each of the sleds 1204 may be
similar to, embodied as, or otherwise form a part of, the sleds
204-1-204-4, 404A-1, 404A-2, 404B-1, 404B-2, 404C-1, 404C-2,
404D-1, 404D-2, 504A, 504B, 704, 1004, 1130, 1132, 1134 described
above. For example, each sled 1204 may be embodied as a compute
sled, a memory sled, an accelerator sled, a data storage sled,
and/or other physical resource sled.
[0054] Each of the illustrative sleds 1204 is housed or otherwise
received by a corresponding server space 1205 of the rack 1202.
When the sleds 1204 are positioned in the server spaces 1205, the
sleds 1204 are spaced from one another in a vertical direction V.
The sleds 1204 illustratively include compute sleds 1204C and a
memory sled 1204M. As discussed in greater detail below, each of
the compute sleds 1204C is communicatively coupled to the memory
sled 1204M by an electromagnetic waveguide 1206.
[0055] Each of the illustrative electromagnetic waveguides 1206
extends between the memory sled 1204M and a corresponding one of
the compute sleds 1204C to communicatively couple the memory sled
1204M to the corresponding compute sled 1204C. Each electromagnetic
waveguide 1206 is sized to span a vertical distance D between the
memory sled 1204M and the corresponding compute sled 1204C.
Compared to other configurations, the use of the waveguides 1206 to
communicatively couple the memory sled 1204M to the compute sleds
1204C may provide one or more benefits. For example, in some
configurations, the vertical distance D between the memory sled
1204M and a corresponding compute sled 1204C may be such that
coupling those components to one another by a conventional
electrical cable or a printed circuit board (PCB) is impractical or
otherwise undesirable. In such configurations, use of the
electromagnetic waveguides 1206 may be desirable. In other
configurations, coupling the memory sled 1204M to a corresponding
compute sled 1204C by a conventional electrical cable or a printed
circuit board may be associated with undesirable cost. In those
configurations, use of the electromagnetic waveguides 1206 may be
preferable.
[0056] Each of the electromagnetic waveguides 1206 illustratively
includes, or is otherwise embodied as, a structure capable of
carrying electromagnetic waves between the sleds 1204. Each
electromagnetic waveguide 1206 is configured to communicate
millimeter wave (MMW) data signals between the memory sled 1204M
and a corresponding compute sled 1204C. In the illustrative
embodiment, each electromagnetic waveguide 1206 is configured to
communicate the millimeter wave data signals between the memory
sled 1204M and the corresponding compute sled 1204C at about 50 to
100 gigabits per second during operation of the rack 1202.
Additionally, in the illustrative embodiment, the millimeter wave
data signals have a carrier frequency range of about 60 to 120
gigahertz.
[0057] Each of the illustrative electromagnetic waveguides 1206
includes a core 1208 that is formed from a dielectric material. In
some embodiments, the core 1208 of each electromagnetic waveguide
1206 may be formed from a solid dielectric material, such as a
polymeric material, porcelain, or glass, for example. The core may
be any suitable dimension, such as a rectangular cross-section with
dimensions of 1.4 by 0.7 millimeters. In other embodiments, the
core 1208 of each electromagnetic waveguide 1206 may be formed from
one another suitable dielectric material, such as a gas or liquid
dielectric material, for example.
[0058] Each of the illustrative electromagnetic waveguides 1206
also includes a metallic coating 1210 that is applied to the core
1208. In some embodiments, the metallic coating 1210 may include,
or otherwise be embodied as, a foil sheath applied to the core
1208. Additionally, in some embodiments, the metallic coating 1210
may include, or otherwise be embodied as, a spray-on coating
applied to the core 1208. However, in other embodiments, the
metallic coating 1210 may omitted from each electromagnetic
waveguide 1206 entirely.
[0059] Each of the illustrative compute sleds 1204C includes one or
more physical resources 1212 and a chassis-less circuit board
substrate 1214, as shown in FIG. 12. The one or more physical
resources 1212 illustratively include, or are otherwise embodied
as, high-power processors. Of course, it should be appreciated that
in other embodiments, the one or more physical resources 1212 may
include, or be otherwise embodied as, accelerator co-processors
storage controllers, and/or network interface controllers, for
example. The circuit board substrate 1214 does not include a
housing or enclosure, which may improve the airflow over the
electrical components of the compute sled 1204C by reducing those
structures that may inhibit air flow. In some embodiments, the
circuit board substrate 1214 may have a geometric shape configured
to reduce the length of the airflow path across the electrical
components mounted to the circuit board substrate 1214.
Additionally, in some embodiments, the various electrical
components mounted to the circuit board substrate 1214 are mounted
in corresponding locations such that no two substantively
heat-producing electrical components shadow each other.
[0060] The compute sleds 1204C illustratively include one set of
compute sleds 1204C-1 and another set of compute sleds 1204C-2. The
set of compute sleds 1204C-2 is arranged above the set of compute
sleds 1204C-1 in the vertical direction V. In the illustrative
embodiment, at least a portion of the memory sled 1204M is arranged
vertically between the sets of compute sleds 1204C-1, 1204C-2.
Additionally, in the illustrative embodiment, each of the sets of
compute sleds 1204C-1, 1204C-2 includes eight compute sleds 1204C.
Of course, it should be appreciated that in other embodiments, the
sets of compute sleds 1204C-1, 1204C-2 may each include another
suitable number of compute sleds 1204C. Furthermore, it should be
appreciated that in other embodiments, the number of compute sleds
1204C included in the set of compute sleds 1204C-1 may be different
from the number of compute sleds 1204C included in the set of
compute sleds 1204C-2.
[0061] The memory sled 1204M illustratively includes one or more
memory devices 1218 and a chassis-less circuit board substrate 1220
that is similar to the circuit board substrate 1214, as shown in
FIG. 12. The one or more memory devices 1218 may act as a far
memory layer of memory in a memory hierarchy between local DRAM on
the compute sled 1204C and storage on, e.g., a storage sled 1132.
The one or more memory devices 1218 may include, or otherwise be
embodied as, any memory device capable of storing data (e.g., the
data carried by millimeter wave data signals) or other information
provided by the physical resources 1212 of the compute sleds 1204C.
For example, the one or more memory devices 1218 may include, or
otherwise be embodied as, dual in-line memory modules (DIMMs),
which may support DDR, DDR2, DDR3, DDR4, or DDR5 random access
memory (RAM). Of course, in other embodiments, the one or more
memory devices 1218 may utilize other memory technologies,
including volatile and/or non-volatile memory. For example, types
of volatile memory may include, but are not limited to, data rate
synchronous dynamic RAM (DDR SDRAM), static random-access memory
(SRAM), thyristor RAM (T-RAM) or zero-capacitor RAM (Z-RAM). Types
of non-volatile memory may include byte or block addressable types
of non-volatile memory. The byte or block addressable types of
non-volatile memory may include, but are not limited to,
3-dimensional (3-D) cross-point memory, memory that uses
chalcogenide phase change material (e.g., chalcogenide glass),
multi-threshold level NAND flash memory, NOR flash memory, single
or multi-level phase change memory (PCM), resistive memory,
nanowire memory, ferroelectric transistor random access memory
(FeTRAM), magnetoresistive random access memory (MRAM) memory that
incorporates memristor technology, or spin transfer torque MRAM
(STT-MRAM), or a combination of any of the above, or other
non-volatile memory types.
[0062] The illustrative rack 1202 includes a connector bank 1222
that is communicatively coupled to the memory sled 1204M. In the
illustrative embodiment, at least a portion of the connector bank
1222 is vertically arranged between the sets 1204C-1, 1204C-2 of
compute sleds 1204. The connector bank 1222 includes one set of
connectors 1222-1 and another set of connectors 1222-2. Each of the
set of connectors 1222-1 illustratively includes, or is otherwise
embodied as, any connector capable of mating with a corresponding
feature (not shown) of a corresponding electromagnetic waveguide
1206. Similarly, each of the set of connectors 1222-2
illustratively includes, or is otherwise embodied as, any connector
capable of mating with a corresponding feature (not shown) of a
corresponding electromagnetic waveguide 1206.
[0063] In the illustrative embodiment, the set of connectors 1222-1
of the connector bank 1222 includes eight connectors 1222-1A,
1222-1B, 1222-1C, 1222-1D, 1222-1E, 1222-1F, 1222-1G, 1222-1H, as
shown in FIG. 12. The connector 1222-1A is configured to mate with
a corresponding feature of the electromagnetic waveguide 1206-1A to
communicatively couple the waveguide 1206-1A to the memory sled
1204M, the connector 1222-1B is configured to mate with a
corresponding feature of the electromagnetic waveguide 1206-1B to
communicatively couple the waveguide 1206-1B to the memory sled
1204M, and so on for connectors 1222-1C to 1222-1H. Corresponding
features of the electromagnetic waveguides 1206-1A to 1206-1H are
configured to mate with corresponding connectors (not shown) of the
compute sleds 1204C-1A to 1204C-1H so that the compute sleds
1204C-1A to 1204C-1H are communicatively coupled to the memory sled
1204M by the waveguides 1206-1A to 1206-1H. Accordingly, in use of
the rack 1202, the waveguides 1206-1A to 1206-1H communicate
millimeter wave data signals between the compute sleds 1204C-1A to
1204C-1H and the memory sled 1204M. In the illustrative embodiment,
the set of connectors 1222-2 of the connector bank 1222 includes
eight connectors 1222-2A, 1222-2B, 1222-2C, 1222-2D, 1222-2E,
1222-2F, 1222-2G, 1222-2H, as shown in FIG. 12, and are configured
in a similar manner as connectors 1222-1, which will not be
repeated in the interest of clarity.
[0064] The illustrative rack 1202 includes a conduit 1224 that
extends vertically between a lowermost sled (i.e., the compute sled
1204C-1A) of the set of compute sleds 1204C-1 and an uppermost sled
(i.e., the compute sled 1204C-1H) of the set of compute sleds
1204C-1, as shown in FIG. 12. The conduit 1224 is coupled to a
central spine 1226 of the rack 1202 that extends vertically. The
conduit 1224 illustratively includes, or is otherwise embodied as,
any structure capable of routing therethrough the electromagnetic
waveguides 1206-1A, 1206-1B, 1206-1C, 1206-1D, 1206-1E, 1206-1F,
1206-1G, 1206-1H. The conduit 1224 may be provided to minimize
disruptions caused by the electromagnetic waveguides 1206-1A,
1206-1B, 1206-1C, 1206-1D, 1206-1E, 1206-1F, 1206-1G, 1206-1H to
airflow over the compute sleds 1204C-1A, 1204C-1B, 1204C-1C,
1204C-1D, 1204C-1E, 1204C-1F, 1204C-1G, 1204C-1H during operation
of the rack 1202.
[0065] The illustrative rack 1202 includes a conduit 1228 that
extends vertically between a lowermost sled (i.e., the compute sled
1204C-2A) of the set of compute sleds 1204C-2 and an uppermost sled
(i.e., the compute sled 1204C-2H) of the set of compute sleds
1204C-2, as shown in FIG. 12. The conduit 1228 is coupled to the
central spine 1226. The conduit 1228 illustratively includes, or is
otherwise embodied as, any structure capable of routing
therethrough the electromagnetic waveguides 1206-2A, 1206-2B,
1206-2C, 1206-2D, 1206-2E, 1206-2F, 1206-2G, 1206-2H. The conduit
1228 may be provided to minimize disruptions caused by the
electromagnetic waveguides 1206-2A, 1206-2B, 1206-2C, 1206-2D,
1206-2E, 1206-2F, 1206-2G, 1206-2H to airflow over the compute
sleds 1204C-2A, 1204C-2B, 1204C-2C, 1204C-2D, 1204C-2E, 1204C-2F,
1204C-2G, 1204C-2H during operation of the rack 1202.
[0066] In some embodiments, each compute sled 1204C of the set of
compute sleds 1204C-1 may be inserted through a side 1230 of the
rack 1202 toward the central spine 1226 as indicated by arrow 1232.
When advanced in the direction indicated by arrow 1232, the
connector of each compute sled 1204C of the set of compute sleds
1204C-1 may be configured to blindly mate with the corresponding
feature of the corresponding electromagnetic waveguide 1206-1A,
1206-1B, 1206-1C, 1206-1D, 1206-1E, 1206-1F, 1206-1G, 1206-1H.
Similarly, in such embodiments, each compute sled 1204C of the set
of compute sleds 1204C-2 may be inserted through the side 1230
toward the central spine 1226 in the direction indicated by arrow
1232. When advanced in the direction indicated by arrow 1232, the
connector of each compute sled 1204C of the set of compute sleds
1204C-2 may be configured to blindly mate with the corresponding
feature of the corresponding electromagnetic waveguide 1206-2A,
1206-2B, 1206-2C, 1206-2D, 1206-2E, 1206-2F, 1206-2G, 1206-2H.
[0067] In some embodiments, the rack 1202 may include more than one
memory sled 1204M. For example, in one embodiment, some or all of
the compute sleds 1204C may have two processors and each compute
sled 1204C is connected to each of two memory sleds 1204M through a
connector bank 1222 in a similar manner as described above. In such
an embodiment, a first processor of each compute sled 1204C may be
configured to primarily or exclusively communicate with a first
memory sled 1204M and a second processor of each compute sled 1204C
may be configured to primarily or exclusively communicate with a
second memory sled 1204M. In another example, a rack 1202 may
include two memory sleds 1204M, each of which is connected to half
of the compute sleds 1204C. In such an embodiment, a first memory
sled 1204M may be positioned in the middle of the top half of the
rack 1202 and connected to the compute sleds 1204C in the top half
of the rack 1202, and a second memory sled 1204M may be positioned
in the middle of the bottom half of the rack 1202 and connected to
the compute sleds 1204C in the bottom half of the rack 1202.
[0068] Referring now to FIGS. 13 and 14, multiple racks 1302A,
1302B may be coupled, mounted, or otherwise situated together to
form a rack pod 1340. The rack pod 1340 may be included in a data
center 1300, which may be substantially similar to the data center
1200. Like the rack 1202, each of the racks 1302A, 1302B is
configured to house or otherwise receive one or more sleds 1304 for
mounting therein. The racks 1302A, 1302B may therefore be
substantially similar to the rack 1202. In addition, each of the
sleds 1304 may be substantially similar to one of the compute sleds
1204C or the memory sled 1204M. Of course, it should be appreciated
that the concepts described in regard to FIGS. 13 and 14 may be
implemented without necessarily implementing the concepts described
above in regard to FIG. 12.
[0069] The rack pod 1340 illustratively includes electromagnetic
waveguides 1406 as shown in FIG. 14. Each of the illustrative
electromagnetic waveguides 1406 extends between a corresponding one
of the sleds 1304 and a switch 1342 that is arranged between the
racks 1302A, 1302B to communicatively couple the corresponding sled
1304 to the switch 1342. Compared to other configurations, the use
of the waveguides 1406 to communicatively couple the sleds 1304 to
the switch 1342 may provide one or more benefits. For example, in
some configurations, the distance between one or more of the sleds
1304 and the switch 1342 may be such that coupling those components
to one another by a conventional electrical cable or a printed
circuit board is impractical or otherwise undesirable, as shown in
FIG. 13. In such configurations, use of the electromagnetic
waveguides 1406 may be desirable. In other configurations, coupling
the sleds 1304 to the switch 1342 by conventional electrical cables
or printed circuit boards may limit the amount of data communicated
between the sleds 1304 and the switch 1342 over a given time
period. In those configurations, use of the electromagnetic
waveguides 1406 may be preferable. Furthermore, use of the
electromagnetic waveguides 1406 may facilitate, or otherwise be
associated with, the use of a switch (i.e., the switch 1342) having
a higher radix than a switch communicatively coupled to the sleds
1304 by conventional electrical cables or printed circuit boards.
Of course, it should be appreciated that the switch 1342 could be
in a different position than what is shown in FIG. 13 such that the
distance between one or more of the sleds 1304 and the switch 1342
may be different. For example, the switch 1342 could be located at
any vertical position between racks, it could be at the top of one
of the racks (such as the center-most racks), or it could be
straddling two racks (such as the two center-most racks).
[0070] The illustrative electromagnetic waveguides 1406 are
substantially similar to the waveguides 1206. As such, each of the
electromagnetic waveguides 1406 illustratively includes, or is
otherwise embodied as, a structure capable of carrying millimeter
wave data signals between a corresponding sled 1304 and the switch
1342. Like each electromagnetic waveguide 1206, each
electromagnetic waveguide 1406 is illustratively configured to
communicate the millimeter wave data signals between the
corresponding sled 1304 and the switch 1342 at about 50 to 100
gigabits per second during operation of the rack pod 1340. In the
illustrative embodiment, the millimeter wave data signals have a
carrier frequency range of about 60 to 120 gigahertz. Furthermore,
like each electromagnetic waveguide 1206, each electromagnetic
waveguide 1406 includes a core 1408 that is formed from a
dielectric material and a metallic coating 1410 that is applied to
the core 1408.
[0071] In the illustrative embodiment, at least one sled 1304 of
each of the racks 1302A, 1302B is communicatively coupled to the
switch 1342 by a corresponding electromagnetic waveguide 1406. To
mate with corresponding features 1506 (see FIG. 15) of the
waveguides 1406 such that the waveguides 1406 are communicatively
coupled to the switch 1342, the switch 1342 includes waveguide
connectors 1444 that are communicatively coupled to a switch chip
1458. In the illustrative embodiment, the number of waveguide
connectors 1444 corresponds to the number of racks 1302 included in
the rack pod 1340. That is, each waveguide connector 1444 is
configured to mate with corresponding features of one or more
electromagnetic waveguides 1406 that are communicatively coupled to
one or more sleds 1304 mounted in a corresponding rack 1302. In
addition to the racks 1302A, 1302B, the illustrative rack pod 1340
includes racks 1402C, 1402D, 1402E, 1402F, 1402G, 1402H, as best
seen in FIG. 14. The illustrative waveguide connectors 1444
therefore include eight connectors 1444A, 1444B, 1444C, 1444D,
1444E, 1444F, 1444G, 1444H to mate with the corresponding features
of respective electromagnetic waveguides 1406A, 1406B, 1406C,
1406D, 1406E, 1406F, 1406G, 1406H that are communicatively coupled
to respective racks 1302A, 1302B, 1402C, 1402D, 1402E, 1402F,
1402G, 1402H. In the illustrative embodiment, each of the
electromagnetic waveguides 1406A, 1406B, 1406C, 1406D, 1406E,
1406F, 1406G, 1406H includes sixteen separate waveguides.
[0072] Each electromagnetic waveguide 1406 is coupled to a
corresponding sled 1304 at a sled coupling point 1350, as shown in
FIG. 13. Each electromagnetic waveguide 1406 is also coupled to the
switch 1342 at a switch coupling point 1352. In the illustrative
embodiment, the Manhattan Distance MD between the sled coupling
point 1350 and the switch coupling point 1352 is no greater than
about 10 feet. Additionally, in the illustrative embodiment, the
shortest distance D between the sled coupling point 1350 and the
switch coupling point 1352 is no greater than about 7.6 feet. Each
electromagnetic waveguide 1406 illustratively extends at least 7
feet between the sled coupling point 1350 and the switch coupling
point 1352.
[0073] The illustrative switch 1342 includes optical connectors
1446 that are supported on a surface 1448 of the switch 1342, as
shown in FIG. 14. The optical connectors 1446 are arranged above
the waveguide connectors 1444 relative to a surface 1450 of the
switch 1342. The optical connectors 1446 illustratively include, or
is otherwise embodied as, any connectors capable of mating with
corresponding connectors (not shown) of spine switches 1452
included in the rack pod 1340 to communicatively couple the switch
1342 to the spine switches 1452. In some embodiments, the optical
connectors 1446 may include, or otherwise by embodied as, quad
small form-factor pluggable (QSFP) connectors, each of which may be
capable of communicating 400 gigabits of data per second. The spine
switches 1452 may be included in a dual-mode optical switching
infrastructure, such as the dual-mode optical switching
infrastructure 514, for example. Accordingly, the spine switches
1452 may include, or otherwise be embodied as, dual-mode optical
switches similar to the dual-mode optical switches 515, multi-ply
switches, dual-mode optical spine switches similar to the dual-mode
optical spine switches 520, or leaf switches similar to the leaf
switches 530.
[0074] In the illustrative embodiment, the optical connectors 1446
include one set of optical connectors 1446A and another set of
optical connectors 1446B. The set of optical connectors 1446A is
arranged below the set of optical connectors 1446B relative to the
surface 1448. Each set of optical connectors 1446A, 1446B includes
16 optical connectors. Of course, it should be appreciated that in
other embodiments, each set of optical connectors 1446A, 1446B may
include another suitable number of optical connectors.
Additionally, in other embodiments, the number of optical
connectors included in the set 1446A may be different from the
number of optical connectors included in the set 1446B.
[0075] The illustrative switch 1342 includes chip connectors 1456
that are communicatively coupled to the switch chip 1458 mounted on
the surface 1450. The chip connectors 1456 illustratively include,
or are otherwise embodied as, any connectors capable of interfacing
with the optical connectors 1446 to communicatively couple the
optical connectors 1446 to the switch chip 1458. The chip
connectors 1456 include eight connectors. However, in other
embodiments, the chip connectors 1456 may include another suitable
number of connectors. The switch chip 1458 provides signal
connectivity between the waveguide connectors 1444 and the optical
connectors 1446. The chip switch 1458 is illustratively configured
to process 25.6 terabits of data per second.
[0076] In the illustrative embodiment, the waveguide connectors
1444A, 1444B, 1444C, 1444D, 1444E, 1444F, 1444G, 1444H of the
switch 1342 may be blindly mated with the connectors 1506 of the
respective electromagnetic waveguides 1406A, 1406B, 1406C, 1406D,
1406E, 1406F, 1406G, 1406H to communicatively couple the waveguide
connectors 1444A, 1444B, 1444C, 1444D, 1444E, 1444F, 1444G, 1444H
to the waveguides 1406A, 1406B, 1406C, 1406D, 1406E, 1406F, 1406G,
1406H. Additionally, in the illustrative embodiment, the optical
connectors 1446 may be blindly mated with the corresponding
connectors of the spine switches 1452 to communicatively couple the
optical connectors 1446 to the corresponding connectors of the
spine switches 1452.
[0077] Referring now to FIG. 15, an illustrative connector 1506 may
include several connector ports 1508, with each connector port 1508
corresponding to one of the waveguides 1406. In such a
configuration, a connector port 1508 may be capable of
communicating 50 to 100 gigabits of data per second, with an
aggregate connectivity of, e.g., 800 to 1,600 gigabits of data per
second for the connector 1506. It should be appreciated that each
of the illustrative waveguides 1406 includes a pair of connectors
1506, one of which is configured to mate with a corresponding sled
1304 and the other of which is configured to mate with a
corresponding waveguide connector 1444. In some embodiments, a
connector 1506 with a large number of connector port 1508 may
connect to several connectors 1506 with a fewer number of connector
ports 1508. For example, a connector 1506 connected to the switch
1342 may have enough ports 1508 to carry 1.6 terabits per second,
and that connector 1506 may be connected by electromagnetic
waveguides 1406 to each of the 16 sleds on a single rack, with a
connector 1506 with enough connector ports 1508 to carry 400
gigabits per second. Additionally, it should be appreciated that
the illustrative connector 1506 may be included in, or otherwise
embodied as, a connector of one or more of the electromagnetic
waveguides 1206 (discussed above) or the electromagnetic waveguides
2006 (discussed below). The illustrative connector ports 1508
include four rows 1508R-1, 1508R-2, 1508R-3, 1508R-4 of connector
ports 1508. Additionally, the illustrative connector ports 1508
include four columns 1508C-1, 1508C-2, 1508C-3, 1508C-4 of
connector ports 1508. However, in other embodiments, the connector
ports 1508 may include another suitable number of rows and
columns.
[0078] Referring now to FIG. 16, an illustrative waveguide
connector 1444 of the switch 1342 is configured to mate with the
connectors 1506 of the electromagnetic waveguides 1406 that are
communicatively coupled to the sleds 1304 mounted in one of the
racks 1302A, 1302B, 1402C, 1402D, 1402E, 1402F, 1402G, 1402H. Using
the rack 1302A as an example, which includes at least sixteen sleds
1304A, the waveguide connector 1444A is configured to mate with the
connectors 1506 of the at least sixteen electromagnetic waveguides
1406A that are each communicatively coupled to one of the at least
sixteen sleds 1304A. Accordingly, the waveguide connector 1444
includes waveguide connector ports 1646 that have at least sixteen
rows 1646R-1, 1646R-2, 1646R-3, 1646R-4, 1646R-5, 1646R-6, 1646R-7,
1646R-8, 1646R-9, 1646R-10, 1646R-11, 1646R-12, 1646R-13, 1646R-14,
1646R-15, 1646R-16 of connector ports 1646. Additionally, the
waveguide connector ports 1646 have at least sixteen columns
1646C-1, 1646C-2, 1646C-3, 1646C-4, 1646C-5, 1646C-6, 1646C-7,
1646C-8, 1646C-9, 1646C-10, 1646C-11, 1646C-12, 1646C-13, 1646C-14,
1646C-15, 1646C-16. However, in other embodiments, the connector
ports 1646 may include another suitable number of rows and columns.
In any case, each of the illustrative connector ports 1646 is
capable of communicating 50 to 100 gigabytes of data.
[0079] Again, using the rack 1302A as an example, when the
connectors 1506 of the at least sixteen electromagnetic waveguides
1406A are mated with the at least sixteen sleds 1304A and with the
waveguide connector 1444A, the waveguide connector 1444A is
communicatively coupled to the at least sixteen sleds 1304A.
Because the switch 1342 includes the eight waveguide connectors
1444A, 1444B, 1444C, 1444D, 1444E, 1444F, 1444G, 1444H, the switch
1342 is connected to at least 128 of the sleds 1304 in use of the
rack pod 1340. The switch 1342 therefore illustratively includes,
or is otherwise embodied as, a high-radix switch. The switch 1342
also may be said to have a radix of 128.
[0080] Referring now to FIG. 17, a connectivity scheme 1700 that
may generally be representative of link-layer connectivity may be
established in some embodiments among various racks 1702 of a rack
pod 1740, which may be included in a data center such as any of
example data centers 100, 300, 400, 1200, and 1300 of respective
FIGS. 1, 3, 4, 12, and 13, for instance. The connectivity scheme
1700 may be connected to or otherwise form a part of an optical
fabric that features a dual-mode optical switching infrastructure,
such as the dual-mode optical switching infrastructure 514. The
connectivity scheme 1700 may incorporate, or otherwise be
implemented with the use of, at least one pod switch 1842 (see FIG.
18). As described in greater detail below, in use of the rack pod
1740, the at least one pod switch 1842 is communicatively coupled
to the racks 1702 and to spine switches 1752 included in the rack
pod 1740. Additionally, as further discussed below, electromagnetic
waveguides 2006 (see FIG. 20) arranged on a waveguide backplane
1908 (see FIG. 19) of the at least one pod switch 1842
communicatively couple the racks 1702 to the spine switches 1752
through the at least one pod switch 1842 in use of the rack pod
1740.
[0081] In the illustrative embodiment, each of the racks 1702 is
configured to house or otherwise receive one or more sleds 1704 for
mounting therein. The racks 1702 are therefore substantially
similar to the racks 1202, 1302A, 1302B. In addition, each of the
sleds 1704 is substantially similar to one of the sleds 1304 and to
one of the compute sleds 1204C or the memory sled 1204M. In some
embodiments, the racks 1702 may include 64 racks. In those
embodiments, each of the 64 racks may include 16 sleds of the sleds
1704. Thus, in those embodiments, the sleds 1704 may include 1024
sleds.
[0082] Compared to other configurations, the use of the
electromagnetic waveguides 2006 to communicatively couple the racks
1702 to the spine switches 1752 through the pod switch 1842 may
provide one or more benefits. For example, in some configurations,
communicatively coupling the racks 1702 to the spine switches 1752
by conventional electrical cables or printed circuit boards may be
impractical or otherwise undesirable from the standpoint of
supplying power and/or transferring data at particular bandwidths
between the racks 1702 and the spine switches 1752. In such
configurations, use of the electromagnetic waveguides 2006 may be
desirable. In other configurations, communicatively coupling the
racks 1702 to the spine switches 1752 by conventional electrical
cables or printed circuit boards may be associated with undesirable
cost. In those configurations, use of the electromagnetic
waveguides 2006 may be preferable.
[0083] Referring now to FIG. 18, a front panel or front side 1808
of the pod switch 1842 that is arranged opposite the waveguide
backplane or back side 1908 illustratively includes a rack
interface 1810 and a spine switch interface 1820. The illustrative
rack interface 1810 includes, or is otherwise embodied as, a set of
line cards 1812 having optical connectors 1814. The optical
connectors 1814 include, or are otherwise embodied as, any
connectors capable of mating with corresponding connectors (not
shown) of the spine switches 1852 to communicatively couple the pod
switch 1842 to the spine switches 1852. The spine switch interface
1820 includes, or is otherwise embodied as, a set of line cards
1822 having optical connectors 1824. The optical connectors 1824
include, or are otherwise embodied as, any connectors capable of
mating with corresponding connectors (not shown) of the sleds 1704
to communicatively couple the pod switch 1842 to the sleds 1704. As
discussed below, in use of the rack pod 1740, each one of the
electromagnetic waveguides 2006 communicatively couples at least
one of the line cards included in the set 1812 to at least one of
the line cards included in the set 1822 to communicate millimeter
wave data signals between the at least one line card of the set
1812 and the at least one line card of the set 1822.
[0084] In the illustrative embodiment, the set of line cards 1822
is arranged above the set of line cards 1812 relative to a support
surface (not shown) on which the pod switch 1842 rests. Of course,
it should be appreciated that in other embodiments, the sets of
line cards 1812, 1822 may have another suitable arrangement
relative to one another. In any case, the set of line cards 1812
illustratively includes eight line cards 1812-1, 1812-2, 1812-3,
1812-4, 1812-5, 1812-6, 1812-7, 1812-8 that are spaced from one
another in a vertical direction indicated by arrow V1. Similarly,
the set of line cards 1822 illustratively includes eight line cards
1822-1, 1822-2, 1822-3, 1822-4, 1822-5, 1822-6, 1822-7, 1822-8 that
are vertically spaced from one another. However, in other
embodiments, each of the sets of line cards 1812, 1822 may include
another suitable number of line cards. Additionally, in other
embodiments, the number of line cards included in the set 1812 may
be different from the number of line cards included in the set
1822.
[0085] Each of the illustrative line cards 1812-1, 1812-2, 1812-3,
1812-4, 1812-5, 1812-6, 1812-7, 1812-8 includes 32 of the optical
connectors 1814, as shown in FIG. 18. In some embodiments, the
optical connectors 1814 may include, or otherwise by embodied as,
quad small form-factor pluggable connectors, each of which may be
capable of communicating between 50 to 400 gigabits of data per
second. Of course, it should be appreciated that in other
embodiments, each of the illustrative line cards 1812-1, 1812-2,
1812-3, 1812-4, 1812-5, 1812-6, 1812-7, 1812-8 may include another
suitable number of the optical connectors 1814.
[0086] Each of the illustrative line cards 1822-1, 1822-2, 1822-3,
1822-4, 1822-5, 1822-6, 1822-7, 1822-8 includes 32 of the optical
connectors 1824, as shown in FIG. 18. In some embodiments, the
optical connectors 1824 may include, or otherwise by embodied as,
quad small form-factor pluggable connectors, each of which may be
capable of communicating between 50 to 400 gigabits of data per
second. Of course, it should be appreciated that in other
embodiments, each of the illustrative line cards 1822-1, 1822-2,
1822-3, 1822-4, 1822-5, 1822-6, 1822-7, 1822-8 may include another
suitable number of the optical connectors 1824.
[0087] Referring now to FIG. 19, the waveguide backplane 1908
illustratively includes, or is otherwise embodied as, waveguide
connectors 1910 that are spaced from one another in a vertical
direction as indicated by arrow V2. More specifically, the
waveguide backplane 1908 includes 16 waveguide connector rows
1910-1, 1910-2, 1910-3, 1910-4, 1910-5, 1910-6, 1910-7, 1910-8,
1910-9, 1910-10, 1910-11, 1910-12, 1910-13, 1910-14, 1910-15,
1910-16 that are vertically spaced from one another. In the
illustrative embodiment, each of the waveguide connector rows
1910-1, 1910-2, 1910-3, 1910-4, 1910-5, 1910-6, 1910-7, 1910-8,
1910-9, 1910-10, 1910-11, 1910-12, 1910-13, 1910-14, 1910-15,
1910-16 includes, or is otherwise embodied as, eight waveguide
connectors 1910. Of course, it should be appreciated that in other
embodiments, the waveguide backplane 1908 may include another
suitable number of waveguide connector rows. Additionally, in other
embodiments, each waveguide connector row may include another
suitable number of waveguide connectors.
[0088] In the illustrative embodiment, the waveguide connector rows
1910-1, 1910-2, 1910-3, 1910-4, 1910-5, 1910-6, 1910-7, 1910-8 are
included in, or otherwise coupled to, the line cards 1812-1,
1812-2, 1812-3, 1812-4, 1812-5, 1812-6, 1812-7, 1812-8.
Additionally, in the illustrative embodiment, the waveguide
connector rows 1910-9, 1910-10, 1910-11, 1910-12, 1910-13, 1910-14,
1910-15, 1910-16 are included in, or otherwise coupled to, the line
cards 1822-1, 1822-2, 1822-3, 1822-4, 1822-5, 1822-6, 1822-7,
1822-8. Each of the illustrative line cards 1812-1, 1812-2, 1812-3,
1812-4, 1812-5, 1812-6, 1812-7, 1812-8, 1822-1, 1822-2, 1822-3,
1822-4, 1822-5, 1822-6, 1822-7, 1822-8 may a switch chip 1958. The
switch chip 1958 is illustratively configured to process 25.6
terabits of data per second.
[0089] Each of the illustrative waveguide connectors 1910 includes,
or is otherwise embodied as, any connector capable of mating with a
corresponding connector (not shown) of one of the electromagnetic
waveguides 2006 to communicatively couple the waveguide connector
1910 to a corresponding electromagnetic waveguide 2006. The
waveguide connector rows 1910-1, 1910-2, 1910-3, 1910-4, 1910-5,
1910-6, 1910-7, 1910-8 are included in, or otherwise correspond to,
a set of waveguide connector rows 1910-S1. The waveguide connector
rows 1910-9, 1910-10, 1910-11, 1910-12, 1910-13, 1910-14, 1910-15,
1910-16 are included in, or otherwise correspond to, a set of
waveguide connector rows 1910-S2. The set of waveguide connector
rows 1910-S2 is arranged vertically above the set of waveguide
connector rows 1910-S1 relative to the support surface on which the
pod switch 1842 rests.
[0090] Referring now to FIG. 20, the waveguide connector row 1910-1
illustratively includes waveguide connectors 2010-1A, 2010-1B,
2010-1C, 2010-1D, 2010-1E, 2010-1F, 2010-1G, 2010-1H. The waveguide
connector rows 1910-9, 1910-10, 1910-11, 1910-12, 1910-13, 1910-14,
1910-15, 1910-16 illustratively include respective waveguide
connectors 2010-9H, 2010-10H, 2010-11H, 2010-12H, 2010-13H,
2010-14H, 2010-15H, 2010-16H. In the illustrative embodiment, the
waveguide connectors 2010-1A, 2010-1B, 2010-1C, 2010-1D, 2010-1E,
2010-1F, 2010-1G, 2010-1H are communicatively coupled to the
respective waveguide connectors 2010-16H, 2010-15H, 2010-14H,
2010-13H, 2010-12H, 2010-11H, 2010-10H, 2010-9H by respective
electromagnetic waveguides 2006-1, 2006-2, 2006-3, 2006-4, 2006-5,
2006-6, 2006-7, 2006-8.
[0091] The waveguide connector row 1910-8 illustratively includes
waveguide connectors 2010-8A, 2010-8B, 2010-8C, 2010-8D, 2010-8E,
2010-8F, 2010-8G, 2010-8H. The waveguide connector rows 1910-9,
1910-10, 1910-11, 1910-12, 1910-13, 1910-14, 1910-15, 1910-16
illustratively include respective waveguide connectors 2010-9A,
2010-10A, 2010-11A, 2010-12A, 2010-13A, 2010-14A, 2010-15A,
2010-16A. In the illustrative embodiment, the waveguide connectors
2010-8A, 2010-8B, 2010-8C, 2010-8D, 2010-8E, 2010-8F, 2010-8G,
2010-8H are communicatively coupled to the respective waveguide
connectors 2010-9A, 2010-10A, 2010-11A, 2010-12A, 2010-13A,
2010-14A, 2010-15A, 2010-16A by respective electromagnetic
waveguides 2006-9, 2006-10, 2006-11, 2006-12, 2006-13, 2006-14,
2006-15, 2006-16.
[0092] The illustrative electromagnetic waveguides 2006 are
substantially similar to the waveguides 1206, 1406. As such, each
of the electromagnetic waveguides 2006-1, 2006-2, 2006-3, 2006-4,
2006-5, 2006-6, 2006-7, 2006-8, 2006-9, 2006-10, 2006-11, 2006-12,
2006-13, 2006-14, 2006-15, 2006-16 illustratively includes, or is
otherwise embodied as, a structure capable of carrying millimeter
wave data signals between a corresponding waveguide connector 1910
of the set of waveguide connector rows 1910-S1 and a corresponding
waveguide connector 1910 of the set of waveguide connector rows
1910-S2 during operation of the pod switch 1842. Like each
electromagnetic waveguide 1206, 1406, each electromagnetic
waveguide 2006-1, 2006-2, 2006-3, 2006-4, 2006-5, 2006-6, 2006-7,
2006-8, 2006-9, 2006-10, 2006-11, 2006-12, 2006-13, 2006-14,
2006-15, 2006-16 is illustratively configured to communicate the
millimeter wave data signals between the corresponding waveguide
connector 1910 of the set of waveguide connector rows 1910-S1 and
the corresponding waveguide connector 1910 of the set of waveguide
connector rows 1910-S2 at about 50 to 100 gigabits per second
during operation of the pod switch 1842. In the illustrative
embodiment, the millimeter wave data signals have a frequency range
of about 60 to 120 gigahertz. Furthermore, like each
electromagnetic waveguide 1206, 1406, each electromagnetic
waveguide 2006 includes a core 2008 that is formed from a
dielectric material and a metallic coating 2012 that is applied to
the core 2008.
Examples
[0093] Illustrative examples of the technologies disclosed herein
are provided below. An embodiment of the technologies may include
any one or more, and any combination of, the examples described
below.
[0094] Example 1 includes a rack comprising a plurality of sleds
vertically spaced from one another, wherein the plurality of sleds
include a memory sled and multiple compute sleds; and a plurality
of electromagnetic waveguides to communicate millimeter wave data
signals, wherein each electromagnetic waveguide communicatively
couples the memory sled to a corresponding one of the multiple
compute sleds.
[0095] Example 2 includes the subject matter of Example 1, and
wherein each electromagnetic waveguide is to communicate the
millimeter wave data signals between the memory sled and the
corresponding one of the multiple compute sleds at about 50 to 100
gigabits per second.
[0096] Example 3 includes the subject matter of any of Examples 1
and 2, and wherein the millimeter wave data signals have a carrier
frequency range of about 60 to 120 gigahertz.
[0097] Example 4 includes the subject matter of any of Examples
1-3, and wherein each electromagnetic waveguide comprises a core
formed from a dielectric material.
[0098] Example 5 includes the subject matter of any of Examples
1-4, and wherein each electromagnetic waveguide comprises a
metallic coating applied to the core.
[0099] Example 6 includes the subject matter of any of Examples
1-5, and wherein the multiple compute sleds comprise a first set of
compute sleds and a second set of compute sleds arranged above the
first set of compute sleds.
[0100] Example 7 includes the subject matter of any of Examples
1-6, and further including a first conduit that extends vertically
between a lowermost sled of the first set of compute sleds and an
uppermost sled of the first set of compute sleds, wherein the first
conduit routes therethrough a first set of electromagnetic
waveguides of the plurality of electromagnetic waveguides that
communicatively couple the memory sled to the first set of compute
sleds to minimize disruptions caused by the first set of
electromagnetic waveguides to airflow over the first set of compute
sleds during operation of the rack.
[0101] Example 8 includes the subject matter of any of Examples
1-7, and further including a second conduit that extends vertically
between a lowermost sled of the second set of compute sleds and an
uppermost sled of the second set of compute sleds, wherein the
second conduit routes therethrough a second set of electromagnetic
waveguides of the plurality of electromagnetic waveguides that
communicatively couple the memory sled to the second set of compute
sleds to minimize disruptions caused by the second set of
electromagnetic waveguides to airflow over the second set of
compute sleds during operation of the rack.
[0102] Example 9 includes the subject matter of any of Examples
1-8, and wherein a portion of the memory sled is arranged
vertically between the first set of compute sleds and the second
set of compute sleds.
[0103] Example 10 includes the subject matter of any of Examples
1-9, and wherein the first and second sets of compute sleds each
comprise eight compute sleds.
[0104] Example 11 includes the subject matter of any of Examples
1-10, and wherein the plurality of electromagnetic waveguides
comprise sixteen electromagnetic waveguides.
[0105] Example 12 includes the subject matter of any of Examples
1-11, and further including a connector bank communicatively
coupled to the memory sled, wherein the connector bank includes a
first set of connectors, and wherein each of the first set of
connectors mates with a corresponding feature of a corresponding
one of the electromagnetic waveguides to communicatively couple a
corresponding one of the first set of compute sleds to the memory
sled.
[0106] Example 13 includes the subject matter of any of Examples
1-12, and wherein the first set of connectors comprises eight
connectors.
[0107] Example 14 includes the subject matter of any of Examples
1-13, and wherein the connector bank includes a second set of
connectors, and wherein each of the second set of connectors mates
with a corresponding feature of a corresponding one of the
electromagnetic waveguides to communicatively couple a
corresponding one of the second set of compute sleds to the memory
sled.
[0108] Example 15 includes the subject matter of any of Examples
1-14, and wherein the second set of connectors comprises eight
connectors.
[0109] Example 16 includes a rack pod comprising a plurality of
racks comprising a first rack and a second rack, wherein each rack
of the plurality of racks has plurality of sleds; a plurality of
electromagnetic waveguides to carry millimeter wave data signals,
wherein the plurality of electromagnetic waveguides include a first
electromagnetic waveguide communicatively coupled to at least one
sled of the plurality of sleds of the first rack and a second
electromagnetic waveguide communicatively coupled to at least one
sled of the plurality of sleds of the second rack; and a switch
arranged between the first rack and the second rack, wherein the
switch includes a plurality of waveguide connectors to mate with
the plurality of electromagnetic waveguides, and wherein the
waveguide connectors include a first waveguide connector to mate
with the first electromagnetic waveguide to communicatively couple
the first electromagnetic waveguide to the switch and a second
waveguide connector to mate with the second electromagnetic
waveguide to communicatively couple the second electromagnetic
waveguide to the switch.
[0110] Example 17 includes the subject matter of Example 16, and
wherein each electromagnetic waveguide carries the millimeter wave
data signals at about 50 to 100 gigabits per second during
operation of the rack pod.
[0111] Example 18 includes the subject matter of any of Examples 16
and 17, and wherein the millimeter wave data signals have a carrier
frequency range of about 60 to 120 gigahertz.
[0112] Example 19 includes the subject matter of any of Examples
16-18, and wherein the plurality of racks comprise eight racks.
[0113] Example 20 includes the subject matter of any of Examples
16-19, and wherein the plurality of waveguide connectors of the
switch comprises eight waveguide connectors.
[0114] Example 21 includes the subject matter of any of Examples
16-20, and wherein each electromagnetic waveguide comprises a core
formed from a dielectric material.
[0115] Example 22 includes the subject matter of any of Examples
16-21, and wherein each electromagnetic waveguide comprises a
metallic coating applied to the core.
[0116] Example 23 includes the subject matter of any of Examples
16-22, and wherein the switch comprises a high-radix switch.
[0117] Example 24 includes the subject matter of any of Examples
16-23, and wherein the switch is connected to 128 sleds of the
plurality of sleds of the plurality of racks.
[0118] Example 25 includes the subject matter of any of Examples
16-24, and wherein each electromagnetic waveguide is coupled to at
least one sled of the plurality of sleds of the plurality of racks
at a sled coupling point and to the switch at a switch coupling
point, and wherein the Manhattan distance between the sled coupling
point and the switch coupling point is no greater than about 10
feet.
[0119] Example 26 includes the subject matter of any of Examples
16-25, and wherein the shortest distance between the sled coupling
point and the switch coupling point is no greater than about 7.6
feet.
[0120] Example 27 includes the subject matter of any of Examples
16-26, and wherein each electromagnetic waveguide is at least 7
feet long.
[0121] Example 28 includes the subject matter of any of Examples
16-27, and wherein (i) each electromagnetic waveguide carries the
millimeter wave data signals at about 50 to 100 gigabits per second
during operation of the rack pod, (ii) each electromagnetic
waveguide is at least 7 feet long, and (iii) the switch has a radix
of least 128.
[0122] Example 29 includes the subject matter of any of Examples
16-28, and wherein the plurality of waveguide connectors of the
switch are blindly mated with the plurality of electromagnetic
waveguides.
[0123] Example 30 includes the subject matter of any of Examples
16-29, and wherein the switch comprises a plurality of optical
connectors to mate with corresponding connectors of spine switches
of the rack pod to communicatively couple the switch to the spine
switches.
[0124] Example 31 includes the subject matter of any of Examples
16-30, and wherein the plurality of optical connectors comprises 32
optical connectors.
[0125] Example 32 includes the subject matter of any of Examples
16-31, and wherein the plurality of optical connectors of the
switch are blindly mated with the corresponding connectors of the
spine switches.
[0126] Example 33 includes a pod switch comprising a first set of
line cards having optical connectors to mate with corresponding
connectors of a plurality of spine switches to communicatively
couple the pod switch to the plurality of spine switches; a second
set of line cards having optical connectors to mate with
corresponding connectors of a plurality of sleds to communicatively
couple the pod switch to the plurality of sleds; and a plurality of
electromagnetic waveguides, wherein each electromagnetic waveguide
communicatively couples at least one line card of the first set of
line cards to at least one line card of the second set of line
cards to communicate millimeter wave data signals between the at
least one line card of the first set of line cards and the at least
one line card of the second set of line cards.
[0127] Example 34 includes the subject matter of Example 33, and
wherein each electromagnetic waveguide carries the millimeter wave
data signals at about 50 to 100 gigabits per second during
operation of the pod switch.
[0128] Example 35 includes the subject matter of any of Examples 33
and 34, and wherein the millimeter wave data signals have a carrier
frequency range of about 60 to 120 gigahertz.
[0129] Example 36 includes the subject matter of any of Examples
33-35, and wherein the first set of line cards is arranged above
the second set of line cards.
[0130] Example 37 includes the subject matter of any of Examples
33-36, and wherein the first set of line cards comprises eight line
cards that are vertically spaced from one another.
[0131] Example 38 includes the subject matter of any of Examples
33-37, and wherein the second set of line cards comprises eight
line cards that are vertically spaced from one another.
[0132] Example 39 includes the subject matter of any of Examples
33-38, and wherein each line card of the first set of line cards
comprises 32 optical connectors.
[0133] Example 40 includes the subject matter of any of Examples
33-39, and wherein each line card of the second set of line cards
comprises 32 optical connectors.
[0134] Example 41 includes the subject matter of any of Examples
33-40, and further including a plurality of waveguide connectors to
mate with the plurality of electromagnetic waveguides, wherein the
first and second sets of line cards are arranged on a first side of
the pod switch and the plurality of waveguide connectors are
arranged on a second side of the pod switch arranged opposite the
first side.
[0135] Example 42 includes the subject matter of any of Examples
33-41, and wherein the plurality of waveguide connectors comprises
16 rows of waveguide connectors that are vertically spaced from one
another.
[0136] Example 43 includes the subject matter of any of Examples
33-42, and wherein each row of waveguide connectors comprises eight
waveguide connectors.
[0137] Example 44 includes the subject matter of any of Examples
33-43, and wherein the rows of waveguide connectors comprise a
first set of waveguide connector rows and a second set of waveguide
connector rows arranged below the first set of waveguide connector
rows.
[0138] Example 45 includes the subject matter of any of Examples
33-44, and wherein each waveguide connector of one row of the
second set of waveguide connector rows is communicatively coupled
to one waveguide connector of each row of the first set of
waveguide connector rows by a corresponding one of the
electromagnetic waveguides.
[0139] Example 46 includes the subject matter of any of Examples
33-45, and wherein each waveguide connector of another row of the
second set of waveguide connector rows is communicatively coupled
to another waveguide connector of each row of the first set of
waveguide connector rows by a corresponding one of the
electromagnetic waveguides.
[0140] Example 47 includes the subject matter of any of Examples
33-46, and wherein each electromagnetic waveguide comprises a core
formed from a dielectric material.
[0141] Example 48 includes the subject matter of any of Examples
33-47, and wherein each electromagnetic waveguide comprises a
metallic coating applied to the core.
* * * * *