• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A host computer for emulating personal computers on respective virtual machines. The host computer includes a plurality of physical computer systems each comprising at least two graphics cards (GC1, GC2, GC2, GC4) and hosts at most 50 virtual machines.
    公开号:FR3047576A1
    申请号:FR1650884
    申请日:2016-02-04
    公开日:2017-08-11
    发明作者:Emmanuel Freund；Asher Criou
    申请人:O Computers；
    IPC主号:
    专利说明:

    HOST COMPUTER TO EMULATE HIGH PERFORMANCE PERSONAL COMPUTERS.
    FIELD OF THE INVENTION
    The present invention relates to the field of dematerialized computing. More specifically, the invention relates to a host computer for emulating high performance personal computers.
    BACKGROUND OF THE INVENTION
    Advances in computer technology have made it more economical for individual users to have their own computer system, which has led to the proliferation of personal computers (PCs). The use of individual PCs is very diverse, ranging from standard desktop operations or Internet browsing to viewing high quality video or video games.
    Continuous advances in this computer technology mean that these personal computers are becoming more powerful but also complex and difficult to manage. For this reason, as well as others more specifically related to resource sharing that allows energy savings for the individual user, there has been a growing interest in the separation of user interface devices, including the screen and keyboard, parts of application processing in the computer system. In this case, the user interface devices are physically located at the user's desktop, while the processing and storage components of the computer are in a remote hosting location.
    The user interface devices then have access, at the host computer, to a dedicated virtual machine via a network (most often the Internet), the virtual machine emulating processing, storage and other resources required.
    UI devices, also known as "client zero" or "thin client", are highly dependent on the host computer that performs the IT functions for the clients. In these configurations, the host computer hosts the operating system and the software applications used by the "client zero" or "thin client", which limits processing resources on the client side.
    It is usual for the host computer to consist of a plurality of physical computer systems (servers), each hosting a plurality of virtual machines. Each virtual machine is connected to a client, and provides a dedicated virtual environment for emulating the functions of a physical personal computer, including processing graphics data to display the screen information on the client screen. The data to be displayed is transferred to the client via the network. The customer has sufficient IT resources to receive and display the data feed. The client also has input / output devices (keyboard, mouse ...) to exchange information or instructions with the virtual machine, via the network. Virtualization allows the creation of as many virtual machines as needed, with the goal of providing all users the same level of service on average, for example, to acquire a web page in less than 100 ms.
    The existing physical computer systems that make up the host computer are essentially configured to perform low-load computing operations (text editing, sending e-mail, desktop management) · for as many clients as possible, in parallel . It appears that this configuration is not suitable to provide a sufficient level of service for high performance applications such as 3D games, high quality video, high quality audio, or 3D graphics tasks. traditionally require the processing power of high-performance personal computers.
    PURPOSE OF THE INVENTION
    An object of the present invention is therefore to provide a host computer configured to emulate high performance personal computers.
    SUMMARY OF THE INVENTION The invention relates to a host computer for emulating personal computers on respective virtual machines, the host computer comprising a plurality of physical computer systems. Each computer system includes at least two graphics cards and hosts at most 50 virtual machines.
    By limiting the number of emulated personal computers on each physical computer system and providing each physical computer system with sufficient resources, and specifically graphical computing resources, it is possible to provide a satisfactory level of service even for intensive graphics applications. .
    According to other non-limiting features of the invention, taken individually or in combination: • each computer system hosts less than 10 virtual machines; • Each graphics card includes a graphics processing unit designed to render complex 2D / 3D graphics in real time; • each computer system includes at least four graphics cards; Each physical computer system includes a CPU capable of executing multimedia instructions; Each computer system comprises a main memory having a latency of less than 14 ns; • Each virtual machine is connected to a remote client over a network.
    FIGURES
    Many other advantages and functions of the present invention will be more apparent after reading the following detailed description, taken in conjunction with the accompanying drawings, in which:
    Figure 1 illustrates a computer architecture for implementing the present invention; Figure 2 illustrates a motherboard architecture of a physical computer system 2 of a host computer, given as an example;
    Figure 3 shows the elements of a video adapter; - Figure 4 shows the elements of a graphics card.
    DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS OF
    THE INVENTION
    In the following description, detailed descriptions of known functions and elements, which could unnecessarily obscure the essence of the present invention, will be omitted.
    In FIG. 1, a computer architecture comprises a host computer 1 which has a plurality of physical computer systems 2. As is well known, the physical computer systems 2 can be configured to host one or a plurality of virtual machine (s) (FIG. s) 3, along with its / their operating system and applications. Virtualization allows a plurality of virtual machines 3 to be hosted in each physical computer system 2 while providing a single physical computer system 2 to a plurality of fully isolated virtual environments. One of the well-known Citrix XenServer virtualization technologies is
    Microsoft Hyper-V, VMware ESXi, Oracle Virtual Box, Quick Emulator Open License GNU (QEMU), etc.
    Each of the virtual machines 1 in the host computer 1 can be dedicated to a specific user. The .... . Users can interact with their dedicated virtual machines 3 from remote clients 4, 4 ', - each being connected to the host computer 1 via a network such as the Internet. Since most, if not all, of the processing is at the host computer 1, the remote clients 4, 4 'may remain very simple, and may include, for example, a simple terminal, a network connector and basic input / output devices (keyboard, mouse ...) as represented by the remote client 4 in FIG. 1. Alternatively, it may be a standard personal computer, with its own Central Unit, its graphics card, peripherals, .... as represented by the remote client 4 '.
    To display the images on the user's terminal, the host computer 1 supplies the remote system 4, 4 'on the network with screen data (and possibly additional data or control data for the input devices / output installed at remote site 21). Conversely, the remote clients 4.4 supply the host computer 1 with command data from the input / output devices installed at the remote site (keyboard, mouse), and possibly other forms of communication. data such as screen and sound data provided by a USB device or integrated into a camera and microphone of the remote client 4 ,, 4 ', or network peripherals, at the remote client, such as printers ... .
    The data exchanged between the host computer 1 and the remote clients 4, 4 'can be compressed to limit the use of the network bandwidth.
    According to the invention, each physical computer system 2 hosts at most fifty virtual machines 3, and preferably less than 10 virtual machines 3. This limitation (together with the architecture of the physical computer system 2 which will be described below) makes it possible to Provide sufficient computing resources for each virtual machine to run high-performance applications with a sufficient level of service. Each virtual machine 10 is created at the time of the connection of the client and comprises a virtual CPU, a virtual main memory, a virtual or physical graphics card as well as other resources. Each virtual machine provides a virtual high performance personal computer that is controlled by a remote client 4,4 '.
    Each physical computer system 2 is configured such that the host computer 1 is capable of emulating high performance personal computers.
    Fig. 2 shows a motherboard architecture of a physical computer system 2 compatible with the invention, given by way of example; The motherboard 5 includes a CPU 6. The CPU 6 represents multimedia instruction sets, and may comprise more than one processor. For example, the CPU 6 is a desktop processor such as an Intel Core ™. The CPU 6 is connected to a main memory 7 via a memory bus 8. The main memory 7 has at least 64 GB and has a low latency of less than 14 ns. The memory bus allows a very high data transfer rate, greater than 1200 Mhz. To do this, the memory bus 7 may consist of gold-plated conductive lines. The CPU 6 is also connected to a set of SB 9 components via a 12th bus. The SB 9 component set connects to devices 11a to 11d, such as hard disks, network controllers, I / O controls ... Data can flow from one device on the motherboard 5 to another via different buses, such as the memory bus 8, the extension bus 12a to 12e, which connects the devices together. Expansion buses 12a to 12e may be of a different nature. For example, the SB 9 component set can connect to a peripheral hard disk 11a via a SATA bus. The SB 9 component set can connect to other 11b devices to lld via PCI or PCIE bus 12b, 12c, 12e. The game of
    components 9 provides the logical interface necessary to allow the transfer of data between the devices 11a to 11d, and between the devices 11a, 11b, 11c, 11d and the CPU 6, and other devices connected to the CPU 6 such the main memory 7. The number of peripherals is not limited to the number indicated in FIG. 2, and the actual physical computer system 2 may have more or less such peripherals, as required.
    According to the invention, each physical computer system 2 of the host computer 1 comprises at least two graphics cards on the motherboard 5. For example, as shown in FIG. 2, the CPU 6 is connected via a bus of graphic card 13 to 4 graphics cards GC1, GC2, GC3, GC4.
    In some examples, a graphics card may be connected to a set of SB 9 components rather than connected directly to the CPU 6, as shown in FIG. This is especially true when this graphics card is dedicated to displaying information at the hosting site. In this case, the graphics card is usually a simple video adapter, allowing to directly connect a monitor to the physical computer system. As shown in Fig. 3, the video adapter 5 typically includes a video memory 31, a conversion circuit 32, and at least one video output port 33. The video adapter usually does not include any dedicated CPU, such as a 2D / 3D accelerator or an audio / video decoder, but the invention does not exclude that the video adapter comprises said dedicated graphics central unit.
    The video memory 31 is provided to temporarily store screen data provided by the CPU 6 or transferred from the main memory 7 (or from other peripherals). The screen data stored in the video memory 31 is processed by the conversion circuit 32 in the appropriate digital or analog format; and supplied to the video output port 33, for example, of the VGA, DVI or HDMI type. The video output port 33 may be connected to a video monitor using appropriate cables (not shown in Fig. 3) to display the image that corresponds to the screen data.
    Referring again to the embodiment shown in FIG. 2, the motherboard 5 of the physical computer system 2 comprises at least two graphic cards GCl, GC2, GC3, GC4, preferably connected directly to the CPU. (that is, without going through a set of SB 9 components). Unlike the video adapter, these graphics cards constitute the computing resources dedicated to the virtual machines that will be operational in the physical computer system 2.
    As shown in FIG. 4, each graphics card GC1, GC2, GC3, GC4 includes a graphics card memory 41 of at least 2 GB, for storing the screen data. It also includes a graphics processing unit (GPU) 42 which receives instructions and data, for example, from the CPU 6, via the graphics card bus 13. The graphics processing unit (GPU) ) 42 processes the received instructions and data and provides the corresponding screen data to the graphics card memory 41. The GPU 42 may also instruct the transfer of the screen data stored in the main memory 7 to the memory of the graphics card 41, via the graphics card bus 13 and the memory bus 8.
    The graphics cards GC1, GC2, GC3, GC4 usually also include a decoder 43 for transforming, for example, coded video (and / or audio) data (such as H.264 files) into screen data for storage in the graphics card memory 41. This encoded video (and / or audio) data may be provided by a DVD player, hard disk, or other device connected to the SB 9 component set.
    A graphics card conversion unit 44 processes the screen data stored in the graphics card memory 41 and sends them to the video output port of the graphics card 45. However, since the screen data is intended to be sent to a remote client 4, 4 'on a network., a conversion unit 44 is not a necessary feature of the graphics card GC1, GC2, GC3, GC4 and will typically not be connected to a display screen. the place of accommodation.
    The graphics card bus 13 connects the CPU 6 directly to the graphics cards GC1, GC2, GC3, GC4 to enable fast data transfer. If the number of connections or channels available on the CPU 6 is not sufficient to connect the desired number of graphics cards, the bus 13 may be equipped with one or a plurality of bus switches 13a which allows allow to share the connections or channels of the CPU 6 in a plurality of graphics cards.
    The graphics cards GC1, GC2, GC3, GC4 on the motherboard 5 do not need to be identical, or of the same type, or the same level of performance (as measured by the number of instructions executed by the GPU 42 per second). For example, in the preferred embodiment shown in FIG. 2, the GC1 graphics card could be of poor performance, the medium performance GC2 graphics card, and the high performance GC3 and GC4 graphics cards.
    In a possible variant of the configuration of the motherboard 5 of the physical computer system 2, the CPU 6 does not include the hardware necessary to connect directly to the main memory 7 and graphics cards GC1, GC2, GC3, GC4. In this case, the motherboard 5 may comprise a set of NB components, respectively connected to the CPU 6, the set of components SB 9, the main memory 7 and the graphics cards GC1, GC2, GC3, GC4. This NB component set redirects data traffic among the various elements that are connected to it.
    A host computer 1 according to the invention, which combines a limited number of virtual machines hosted by each computer system 2 and the presence of at least two graphics cards GC1, GC2 in each computer system 2, allows easy execution of high applications. performance, and emulate, from a remote client perspective, high-performance personal computers.
    权利要求:
    Claims (7)
    [1" id="c-fr-0001]
    A host computer (1) for emulating personal computers on respective virtual machines (3), the host computer (1) comprising a plurality of physical computer systems (2) characterized in that each computer system (2) comprises minus two graphics cards (GC1, GC2, GC2, GC4) and hosts at most 50 virtual machines (3).
    [2" id="c-fr-0002]
    2. Host computer (1) according to claim 1, characterized in that each physical computer system (2) hosts less than 10 virtual machines (3).
    [3" id="c-fr-0003]
    3. Host computer (1) according to any one of the preceding claims, characterized in that each graphics card (GC1, GC2, GC2, GC4) comprises a graphics processing unit (44) designed to render 2D / 3D complex graphics in real time.
    [4" id="c-fr-0004]
    4. Host computer (1) according to any one of the preceding claims, characterized in that each physical computer system (2) includes at least four graphics cards (GC1, GC2, GC2, GC4).
    [5" id="c-fr-0005]
    5. host computer (1) according to any one of the preceding claims, characterized in that each physical computer system (2) includes a CPU (6) capable of executing multimedia instructions.
    [6" id="c-fr-0006]
    6. Host computer (1) according to any one of the preceding claims, characterized in that each physical computer system includes a main memory having a latency of less than 14 ns.
    [7" id="c-fr-0007]
    7. Host computer (1) according to any one of the preceding claims, characterized in that each virtual machine (3) is connected to a remote client (4, 4 ') via a network.
    类似技术:
    公开号 | 公开日 | 专利标题
    FR3047576A1|2017-08-11|
    US8872835B2|2014-10-28|Prevention of DoS attack by a rogue graphics application
    Herrera2014|NVIDIA GRID: Graphics accelerated VDI with the visual performance of a workstation
    US9479570B2|2016-10-25|System and method for processing load balancing of graphic streams
    US9537949B2|2017-01-03|Computer system image clustering architecture and use
    US9626166B1|2017-04-18|Common secure cloud appliance image and deployment
    FR3047579A1|2017-08-11|
    US20120262489A1|2012-10-18|Relative and Absolute Screen Rotation Draft Agent
    US8924700B1|2014-12-30|Techniques for booting from an encrypted virtual hard disk
    US20120050260A1|2012-03-01|Method and apparatus for accommodating display migration among a plurality of physical displays
    TWI631504B|2018-08-01|Multi architecture manager
    US9715749B2|2017-07-25|Display of graphical representations of legends in virtualized data formats
    US9839843B1|2017-12-12|Coordination of content presentation operations
    Scroggins2017|Emerging virtualization technology
    US20200036816A1|2020-01-30|Application sharing
    FR3039906A1|2017-02-10|METHOD FOR INTERACTIVE SHARING OF APPLICATIONS AND DATA BETWEEN TOUCHSCREEN COMPUTERS AND COMPUTER PROGRAM FOR IMPLEMENTING SAID METHOD
    EP3311565B1|2020-09-09|Low latency application streaming using temporal frame transformation
    EP3301574B1|2021-11-10|Method for managing graphic cards in a computing system
    US9929485B2|2018-03-27|Card edge connector using a set of electroactive polymers
    US20140015842A1|2014-01-16|Implementing previously rendered frame buffer information in a customized gui display
    KR101430729B1|2014-08-14|Server and method for supporting software
    Li et al.2017|The implementation of a GPU-accelerated virtual desktop infrastructure platform
    Wang et al.2013|A strategy to move taiwan's it industry from commodity hardware manufacturing to competitive cloud solutions
    Mwakisole et al.2019|Cloud computing architecture for elearning systems in secondary schools in Tanzania
    Schulz2017|Software-Defined Data Infrastructure Essentials: Cloud, Converged, and Virtual Fundamental Server Storage I/O Tradecraft
    同族专利:
    公开号 | 公开日
    FR3047576B1|2020-05-01|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20110050712A1|2009-08-26|2011-03-03|Red Hat, Inc.|Extension To A Hypervisor That Utilizes Graphics Hardware On A Host|
    US20150009222A1|2012-11-28|2015-01-08|Nvidia Corporation|Method and system for cloud based virtualized graphics processing for remote displays|FR3077906A1|2018-02-09|2019-08-16|Blade|METHOD FOR BROADCASTING USER SESSIONS, HOST COMPUTER FOR CARRYING OUT SAID METHOD, AND METHOD FOR PROVIDING SERVICE TO A USER|
    FR3086425A1|2018-09-20|2020-03-27|Blade|METHOD FOR EXECUTING AN APPLICATION PROGRAM IN A COMPUTER SYSTEM|
    EP3654613A1|2018-11-16|2020-05-20|Blade|Protocol for transmitting a data flow in transit between a host computer and a remote client|
    EP3654181A1|2018-11-16|2020-05-20|Blade|Method for maintenance of a virtual machine hosted on a server of a host computer|
    FR3090945A1|2018-12-24|2020-06-26|Blade|Method of connecting a remote device to a virtual local area network|
    FR3090946A1|2018-12-24|2020-06-26|Blade|Portable adapter suitable for virtual mounting of a remote device on a virtual machine, system and method.|
    US11012479B2|2018-11-16|2021-05-18|Blade|Methods for capturing and broadcasting user computer sessions and related computer program products|
    法律状态:
    2017-05-23| PLFP| Fee payment|Year of fee payment: 2 |
    2017-08-11| PLSC| Publication of the preliminary search report|Effective date: 20170811 |
    2018-01-19| CD| Change of name or company name|Owner name: BLADE, FR Effective date: 20171218 |
    2018-02-01| PLFP| Fee payment|Year of fee payment: 3 |
    2020-02-28| PLFP| Fee payment|Year of fee payment: 5 |
    2021-02-11| PLFP| Fee payment|Year of fee payment: 6 |
    2022-02-15| PLFP| Fee payment|Year of fee payment: 7 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1650884|2016-02-04|
    FR1650884A|FR3047576B1|2016-02-04|2016-02-04|HOST COMPUTER FOR EMULATING HIGH PERFORMANCE PERSONAL COMPUTERS|FR1650884A| FR3047576B1|2016-02-04|2016-02-04|HOST COMPUTER FOR EMULATING HIGH PERFORMANCE PERSONAL COMPUTERS|
    [返回顶部]