专利汇可以提供MULTIPLE GRAPHICS PROCESSING UNIT PLATFORM专利检索,专利查询,专利分析的服务。并且A multiple graphics processing unit (multi-GPU) platform including a cubby chassis and at least one multi-GPU sled. The cubby chassis includes partitions defining a plurality of sled positions. The multi-GPU sled includes a chassis having a vertical sidewall and a horizontal bottom wall with an open top and an open side. A side-plane PCB is mounted to the vertical sidewall and a plurality of dividers are attached to the bottom wall and oriented perpendicular to the side-plane PCB. One or more GPU cards are connected to the side-plane PCB and are supported on the plurality of dividers. The GPU cards include a GPU PCB having a first side facing the bottom wall and an outward facing second side. A cover is coupled to the horizontal bottom wall to enclose the open side of the sled chassis and help direct airflow across the GPU cards.,下面是MULTIPLE GRAPHICS PROCESSING UNIT PLATFORM专利的具体信息内容。
What is claimed is:
This application is a continuation-in-part of U.S. Non-Provisional application Ser. No. 15/063,024, filed Mar. 7, 2016, which claims priority to and the benefit of U.S. Provisional Application No. 62/129,592, filed Mar. 6, 2015, the disclosures of which are hereby incorporated by reference in their entireties.
This patent application is directed to a multiple graphics processing unit platform with a modular chassis configuration.
Existing server platforms generally include a single printed circuit board (PCB) with multiple servers. Thus, if one server fails, the entire PCB must be replaced. In addition, the server density is not always maximized for a given form factor. Similarly, conventional graphics processing unit (GPU) cards are typically very large which limits their density for a given form factor. Therefore, there is a need for easily serviceable server and GPU platforms that use space efficiently and maximizes the performance of the servers and GPUs.
The disaggregated device slide for parallel graphics processing introduced here may be better understood by referring to the following Detailed Description in conjunction with the accompanying drawings and appendices, all of which are incorporated herein and constitute a part of this specification:
The headings provided herein are for convenience only and do not necessarily affect the scope of the embodiments. Further, the drawings have not necessarily been drawn to scale. For example, the dimensions of some of the elements in the figures may be expanded or reduced to help improve the understanding of the embodiments. Moreover, while the disclosed technology is amenable to various modifications and alternative forms, specific embodiments have been shown by way of example in the drawings and are described in detail below. The intention, however, is not to limit the embodiments described. On the contrary, the embodiments are intended to cover all modifications, equivalents, and alternatives falling within the scope of this disclosure.
A multi-node server platform is disclosed that includes a modular chassis design including a cubby chassis housing multiple multi-server sleds. Each sled includes multiple modular server cards. In some embodiments, the platform has an agnostic architecture facilitated by an integrated circuit that performs a “bridging function” between each server's central processing unit (CPU) and a single 10G communication connector located on the front of each sled.
Each sled houses multiple server cards (e.g., four servers) in a form factor that allows three sleds to be mounted side-by-side in a corresponding cubby. The sled includes a sturdy chassis supporting a vertically oriented side plane PCB. The server cards are inserted horizontally into the side plane PCB and are supported by edge guides disposed on dividers mounted to the chassis. In some embodiments, the dividers are plastic and include large air flow openings. The server cards are mounted in the sled such that the components (e.g., memory and heat sink) extend downward into the sled, the server card encloses the top of the chassis for proper airflow, and the server connectors are positioned closer to the sled's network connection for improved signal integrity. The sled also includes air duct covers to enclose the side of the chassis and direct the airflow through the sled. The cubby chassis provides power to each sled by a single floating bus bar connector that splits the power connections into three sets, each of which connects to a corresponding sled.
In an embodiment, the multi-node server platform includes a cubby chassis and at least one multi-server sled. The cubby chassis includes one or more partitions defining a plurality of sled positions. The multi-server sled is located in a corresponding one of the plurality of sled positions. The multi-server sled includes a sled chassis having a vertical sidewall and a horizontal bottom wall, wherein the sled chassis has an open top and an open side opposite the vertical sidewall. A vertically oriented side-plane PCB is mounted to the vertical sidewall and a plurality of dividers are attached to the bottom wall and are oriented substantially perpendicular to the side-plane PCB. One or more server cards are connected to the side-plane PCB and are horizontally supported by a pair of the plurality of dividers. The server cards include a server PCB having a first side facing the bottom wall and an outward facing second side enclosing the open top of the sled chassis with server components coupled to the first side of the server PCB. A cover is coupled to the horizontal bottom wall to enclose the open side of the sled chassis and help direct airflow across the server cards.
A multi-GPU sled is disclosed. Each sled houses multiple graphics processing units (multi-GPU) cards (e.g., four GPUs) in a PCI form factor that allows three sleds to be mounted side-by-side in a corresponding cubby. A mini-server manages the GPUs and interfaces with a network interface card. The sled includes a sheet metal chassis supporting a vertically oriented sideplane PCB. The GPU cards are inserted horizontally into the sideplane PCB and are supported on by edge guides disposed on dividers mounted to the chassis. In some embodiments, the sled also includes an air duct cover to enclose the side of the chassis and direct the airflow through the sled.
Although the platform may be described herein with respect to selected multiples of components, sleds, server modules, graphical processor units, and cubby positions, for example, it should be understood that other multiples may be implemented.
Various examples of the devices introduced above will now be described in further detail. The following description provides specific details for a thorough understanding and enabling description of these examples. One skilled in the relevant art will understand, however, that the techniques and technology discussed herein may be practiced without many of these details. Likewise, one skilled in the relevant art will also understand that the technology can include many other features not described in detail herein. Additionally, some well-known structures or functions may not be shown or described in detail below so as to avoid unnecessarily obscuring the relevant description.
The terminology used below is to be interpreted in its broadest reasonable manner, even though it is being used in conjunction with a detailed description of some specific examples of the embodiments. Indeed, some terms may even be emphasized below; however, any terminology intended to be interpreted in any restricted manner will be overtly and specifically defined as such in this section.
As shown in
With further reference to
As shown in
As shown in
As can be appreciated in
The server cards 300 are supported by card guides 226, shown in
As shown in
The cubby chassis and sled platform described above with respect to multi-node servers is applicable to graphics processing unit (GPU) cards as well. The multi-GPU sled described below packages multiple GPU cards into a sled in a similar fashion as that described above. Each GPU card uses standard GPU components mounted on a standard peripheral component interconnect (PCI) form factor PCB, which allows for high-density packaging of the GPUs.
As shown in
Referring to
The GPU sled 400 also includes an air duct cover 450 enclosing the otherwise open side of the sled chassis 402 in order to contain the airflow generated by cooling fans 422 within the GPU sled 400. In some embodiments, the air duct cover 450 clips onto a pair of cover pins 458.
The above description, drawings, and appendices are illustrative and are not to be construed as limiting. Numerous specific details are described to provide a thorough understanding of the disclosure. However, in some instances, well-known details are not described in order to avoid obscuring the description. Further, various modifications may be made without deviating from the scope of the embodiments.
Reference in this specification to “one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the disclosure. The appearances of the phrase “in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Moreover, various features are described which may be exhibited by some embodiments and not by others. Similarly, various requirements are described which may be requirements for some embodiments but not for other embodiments.
The terms used in this specification generally have their ordinary meanings in the art, within the context of the disclosure, and in the specific context where each term is used. It will be appreciated that the same thing can be said in more than one way. Consequently, alternative language and synonyms may be used for any one or more of the terms discussed herein, and any special significance is not to be placed upon whether or not a term is elaborated or discussed herein. Synonyms for some terms are provided. A recital of one or more synonyms does not exclude the use of other synonyms. The use of examples anywhere in this specification, including examples of any term discussed herein, is illustrative only and is not intended to further limit the scope and meaning of the disclosure or of any exemplified term. Likewise, the disclosure is not limited to various embodiments given in this specification. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains. In the case of conflict, the present document, including definitions, will control.
标题 | 发布/更新时间 | 阅读量 |
---|---|---|
PLL电路 | 2020-05-08 | 2 |
使用锁相环和锁频环对压控振荡器进行校准以修整其增益 | 2020-06-02 | 1 |
低抖动、宽操作频带及适合低电压操作的频率合成系统 | 2021-07-19 | 2 |
一种宽锁定范围的注入锁定分频器 | 2020-12-07 | 1 |
太阳能电源注锁功率合成高压钠灯 | 2020-11-06 | 3 |
一种灌区水渠液位测量装置 | 2023-01-03 | 2 |
Phase locked loop for controlling an optical recording device and method thereof | 2022-06-13 | 0 |
Phase locked loop provided with digital control frequency multiple oscillator | 2023-12-29 | 2 |
FREQUENCY DIVISION CORRECTION CIRCUIT, RECEPTION CIRCUIT, AND INTEGRATED CIRCUIT | 2020-07-05 | 1 |
FRACTIONAL FREQUENCY DIVIDER | 2023-03-22 | 1 |
高效检索全球专利专利汇是专利免费检索,专利查询,专利分析-国家发明专利查询检索分析平台,是提供专利分析,专利查询,专利检索等数据服务功能的知识产权数据服务商。
我们的产品包含105个国家的1.26亿组数据,免费查、免费专利分析。
专利汇分析报告产品可以对行业情报数据进行梳理分析,涉及维度包括行业专利基本状况分析、地域分析、技术分析、发明人分析、申请人分析、专利权人分析、失效分析、核心专利分析、法律分析、研发重点分析、企业专利处境分析、技术处境分析、专利寿命分析、企业定位分析、引证分析等超过60个分析角度,系统通过AI智能系统对图表进行解读,只需1分钟,一键生成行业专利分析报告。