Patent Application
20130290546
The field relates generally to computing devices and, more particularly, to employing a mechanism for employing and facilitating dynamic and remote memory collaboration at computing devices.
With every new software generation, software applications' memory footprints are growing exponentially, such as outpacing the growth in the capacity of current memory systems (e.g., Random-Access Memory (RAM)). This typically causes an operating system (OS) to start paging in and out of various memory disks (e.g., Hard Disk Drive (HDD), Solid State Drive (SDD), etc.) which operate at several orders of magnitude slower than RAM. Improving performance with larger local memory is not without excessive cost and power implications. Further, current memory technologies usually operate on static configurations and are incapable of handling dynamic changing of the workload conditions.
Embodiments of the present invention are illustrated by way of example and not by way of limitation in the figures of the accompanying drawings, in which like references indicate similar elements and in which:
Embodiments of the invention provide a mechanism for facilitating dynamic and remote memory collaboration at computing devices according to one embodiment of the invention. A method of embodiments of the invention includes dynamically classifying a computing device of a plurality of computing devices as a memory server, where the plurality of computing devices are coupled to each other over a network. The method may further include offering, by the memory server, of memory to be used by one or more of the plurality of computing devices classified as one or more memory clients, and remotely granting, by the memory server, of the memory to the one or more memory clients.
In one embodiment, various computing nodes (e.g., computing devices) in a cluster and/or within a network are enabled to dynamically discover, allocate, and de-allocate remote memory to each other. This technique greatly increases performance, energy efficiency and reduces memory provisioning costs, etc. For example, the nodes may be connected to each other through a communication network (e.g. data center network/intranet, cloud computing, Internet, etc.) or network interconnect (e.g., Ethernet, Infiniband®, Light Peak®, etc.), allowing them to communicate and access remote memory. It a network setting of computing nodes, some computing nodes at certain time periods can have plenty of free memory and that other nodes that are short of memory could use. In one embodiment, a memory collaboration mechanism is employed and run at computing nodes that are available to participate in memory sharing. The memory collaboration mechanism, as will be further described with reference to the subsequent figures, allows for monitoring of local memory dynamics (e.g., using a node classification algorithm), communicate memory requirements with remote computing nodes, and dynamically coordinate memory sharing among computing nodes based on run-time conditions to facilitate remote memory access to achieve, for example, better performance and energy efficiency.
Computing device 100 includes an operating system 106 serving as an interface between any hardware or physical resources of the computer device 100 and a user. Computing device 100 further includes one or more processors 102, memory devices 104, network devices, drivers, or the like, as well as input/output sources, such as touchscreens, touch panels, touch pads, virtual or regular keyboards, virtual or regular mice, etc. It is to be noted that terms like “node”, “computing node”, “client”, “memory client”, “server”, “memory server”, “machine”, “device”, “computing device”, “computer”, “computing system”, “cluster based computer”, and the like, are used interchangeably and synonymously throughout this document.
If, for example, memory of a computing device dips below a minimum memory threshold, the computing device may be automatically classified as a memory client (e.g., a computing device that is short of memory and willing to obtain it) and conversely, if the computing device's memory is greater than a maximum memory threshold, the computing device may be automatically regarded as a memory server (e.g., a computing device that has excessive memory and is willing to share it). There may be other multiple thresholds that can automatically classify a computing device in a neutral zone (e.g., neither memory client nor memory server), such as when the memory amount is greater than the memory minimum high threshold, but lower than the memory maximum low threshold. These thresholds can be modified, added, removed or manipulated in anyway, as necessitated or desired, by the user or dynamically or automatically using one or more control algorithms. In other words, classification of computing devices (or nodes) can be performed dynamically or automatically with each computing device's changing memory amount that corresponding to the changing workload being processed by that computing device, so a memory client can, dynamically, become a memory server and vice versa or remain somewhere in the neutral zone or no change zone based on the available memory amount corresponding to the pending and/or anticipated workload.
The memory collaboration mechanism 108 further includes a memory acquisition protocol 212 whose components 214, 216, 218, 220, 222 provides various functionalities, such as a memory server offers its memory using an offer module 214. The memory server, upon receiving a request for memory from a memory client, grants its memory to the memory client using a grant module 218. The memory client requests the memory using a request module 216 and once the memory is received, the memory client acknowledges the received memory through an acknowledge module 220. The acknowledgment is realized by the memory server through its own acknowledge module 220. In one embodiment, a memory server, using the offer module 214, may broadcast its memory offer to any number of memory clients in a cluster so they may all know that the memory server has a particular amount of memory to offer to any memory client with interested in receiving it. In another embodiment, the memory offer, using the offer module 214, may be sent from the memory server to one or more memory clients (on a one-on-one direct communication basis as opposed to broadcasting the memory offer to all memory clients) based on the knowledge previously communicated to the memory server regarding the one or more memory clients seeking additional memory. This knowledge may be based on the memory requests received at the memory server from the one or more memory clients using the request module 216. As with the memory server, a memory client may also, using the request module 216, either broadcast its memory request to receive additional memory to any number of memory servers in the cluster or place a memory request with individual memory servers based on some previous knowledge, such as a memory offer broadcast by a particular memory server.
In one embodiment, the memory collaboration mechanism 108 further provides a reclaim module 222 so that if and when a memory server is reclassified as a memory client, in needs of memory, it can reclaim the memory previously granted to one or more memory clients. Further, a communication module 232 is employed by the memory collaboration mechanism 108 to facilitate bi- and multi-directional communication between memory servers and memory clients to perform the aforementioned functionalities and tasks (e.g., offering memory, requesting memory, granting memory, receiving memory, acknowledging receipt of memory, reclaiming previously granted memory, etc.). The communication module 232 may be associated with and use any number and types of interfaces and interconnects, such as Light Peak, Infiniband, Ethernet, Network Interface Controller (NIC), and the like.
It is contemplated that any number and type of components may be added to and removed from the memory collaboration mechanism 108 to facilitate dynamic and remote sharing of memory between computing devices. For brevity, clarity, ease of understanding and to focus on the memory collaboration mechanism 108, many of the standard or known components of a computing device are not shown or discussed here.
In the illustrated embodiment and for example, computing device 300A is classified as memory client that employs one or more software applications or programs, an operating system (as the one shown in
In one embodiment, as aforementioned, communication (including granting and/or receiving of memory, transmitting and receiving of memory requests and acknowledgment messages, etc.) between the computing devices 300A, 300B, 300C may be facilitated by one or more communication modules of the one or more memory collaboration mechanisms 108A, 108B, 108C, 108D and through the interconnect (e.g., Light Peak, etc.) and paging and/or extending components employed at these computing devices 300A, 300B, 300C. For example, any communication between the memory client 300A and the memory server 300C may be performed through their respective paging components and the interconnect 332; similarly, communication between the memory client 300A and the memory server 300B may be performed through their respective extending components and the interconnect 332. Further, the three computing devices 300 may be in communication with and/or have access to storage 334 via the interconnect 332.
Now referring to
In one embodiment, a user may predetermine and set a number of memory amount thresholds 412, 414, 416, 418 to influence the classification of a computing device as a memory client 402 or a memory server 404. These thresholds 412, 414, 416, 418 may be set based on user desire or system needs, such as the types of software applications running on the computer device, etc., of the computing device. In the illustrated embodiment and for example, if the computing device's amount of memory falls below a threshold memory amount of low minimum 412, the computing device classified (or reclassified) as memory client 402 as determined by the classification module of the memory collaboration mechanism as shown in
In one embodiment, the classification module also considers the high-end, maximum memory thresholds 416, 418. For example, if the computing device remains below the low maximum 416 amount of memory (but higher than the high minimum threshold 414), the computing device is promoted to neutral 408 (e.g., it is neither classified as memory client 402 nor as memory server 404). On the contrary, in one embodiment, if the computing device gains memory above the high maximum threshold 418, the computing device is classified (or reclassified) as memory server 404. Similar to the low and high minimum thresholds 412, 414, if the computing device's memory amount falls (or rises) between the low and high maximum thresholds 416, 418, the memory server 404 experience no change 410.
In one embodiment, once classified, memory server 422 offers 426 its memory for use by any number of available memory clients. The memory server 422 may broadcast its intention to offer memory to a number of memory clients by broadcasting the intention or offer 426 using, for example, the offer module of the memory collaboration mechanism of
In one embodiment, client memory 444 broadcasts a memory request 446 for memory to any number of memory servers using the request module of the memory collaboration mechanism of
Method 460 begins with block 462 with a computing device of a cluster of computing devices connected or in communication with each other over a network being classified as memory server. At block 464, the classified memory server offers memory to any number of computing devices classified as memory clients. At block 466, requests for memory are received at the memory server from one or more of the memory clients. At block 468, the memory server grants memory to the one or more memory clients. The memory server than receives acknowledgement messages from the one or more memory clients at block 470.
In one embodiment, at block 472, a determination is made as to whether the memory server by reclassified as memory client (or, e.g., neutral, etc.). If the memory server remains as memory server, nothing may change or the process may continue with the memory server offering more memory at block 464. If the memory server is reclassified as a memory client, the process may continue with another determination as to whether the reclassified memory client requires or needs memory at block 474. If not, nothing may change or the process may continue at block 474. If yes, at block 474, the reclassified memory client reclaims the previously granted memory to the one or more memory clients at block 476. The reclaiming may be performed using the reclaim module 222 of the memory collaboration mechanism of
Method 480 begins at block 482 with classification of a computing device of a cluster of computing devices connected over a network as memory client. At block 484, the memory client broadcasts it request for memory to computing devices of the cluster classified as memory servers. At block 486, the memory client receives memory from and granted by one or more memory servers. At block 488, the memory client sends acknowledgements messages to the one or more memory servers that granted it their memory. At block 490, a determination is made as to whether the memory client be reclassified as memory server. If not, the process may remain unchanged or continue with requesting more memory at block 484. If yes, at block 492, the reclassified memory server may release the previously received memory back to the one or more granting memory servers. Further, at block 494, the newly reclassified memory server may offer its excess memory to any number of memory clients.
The computer system 500 further includes a main memory 504, such as a RAM or other dynamic data storage device, coupled to the bus 501 for storing information and instructions to be executed by the processor 502. The main memory also may be used for storing temporary variables or other intermediate information during execution of instructions by the processor. The computer system 500 may also include a nonvolatile memory 506, such as a Read-Only Memory (ROM) or other static data storage device coupled to the bus 501 for storing static information and instructions for the processor.
A mass memory 507 such as a magnetic disk, optical disc, or solid state array and its corresponding drive may also be coupled to the bus 501 of the computer system 500 for storing information and instructions. The computer system 500 can also be coupled via the bus to a display device or monitor 521, such as a Liquid Crystal Display (LCD) or Organic Light Emitting Diode (OLED) array, for displaying information to a user. For example, graphical and textual indications of installation status, operations status and other information may be presented to the user on the display device 521, in addition to the various views and user interactions discussed above.
Typically, user input devices 522, such as a keyboard with alphanumeric, function and other keys, etc., may be coupled to the bus 501 for communicating information and command selections to the processor 502. Additional user input devices 522 may include a cursor control input device such as a mouse, a trackball, a trackpad, or cursor direction keys can be coupled to the bus for communicating direction information and command selections to the processor 502 and to control cursor movement on the display 521.
Camera and microphone arrays 523 are coupled to the bus 501 to observe gestures, record audio and video and to receive visual and audio commands as mentioned above.
Communications interfaces 525 are also coupled to the bus 501. The communication interfaces may include a modem, a network interface card, or other well-known interface devices, such as those used for coupling to Ethernet, token ring, or other types of physical wired or wireless attachments for purposes of providing a communication link to support a LAN or WAN, for example. In this manner, the computer system 500 may also be coupled to a number of peripheral devices, other clients, or control surfaces or consoles, or servers via a conventional network infrastructure, including an Intranet or the Internet, for example.
It is to be appreciated that a lesser or more equipped system than the example described above may be preferred for certain implementations. Therefore, configuration of the computing system 500 may vary from implementation to implementation depending upon numerous factors, such as price constraints, performance requirements, technological improvements, or other circumstances.
Embodiments may be implemented as any or a combination of: one or more microchips or integrated circuits interconnected using a parent-board, hardwired logic, software stored by a memory device and executed by a microprocessor, firmware, an application specific integrated circuit (ASIC), and/or a field programmable gate array (FPGA). The term “logic” may include, by way of example, software or hardware and/or combinations of software and hardware, such as firmware.
Embodiments may be provided, for example, as a computer program product which may include one or more machine-readable media, such as a non-transitory machine-readable medium, having stored thereon machine-executable instructions that, when executed by one or more machines such as a computer, such as computing system 500, network of computers, or other electronic devices, may result in the one or more machines carrying out operations in accordance with embodiments of the present invention. A machine-readable medium may include, but is not limited to, floppy diskettes, optical disks, Compact Disc-ROMs (CD-ROMs), and magneto-optical disks, ROMs, RAMs, Erasable Programmable Read-Only Memories (EPROMs), EEPROMs Electrically Erasable Programmable Read-Only Memories (EEPROMs), magnetic or optical cards, flash memory, or other type of media/machine-readable medium suitable for storing machine-executable instructions, such as solid state storage devices, fast and reliable DRAM sub-systems, etc.
Moreover, embodiments may be downloaded as a computer program product, wherein the program may be transferred from a remote computer (e.g., a server) to a requesting computer (e.g., a client) by way of one or more data signals embodied in and/or modulated by a carrier wave or other propagation medium via a communication link (e.g., a modem and/or network connection). Accordingly, as used herein, a machine-readable medium may, but is not required to, comprise such a carrier wave.
References to “one embodiment”, “an embodiment”, “example embodiment”, “various embodiments”, etc., indicate that the embodiment(s) of the invention so described may include particular features, structures, or characteristics, but not every embodiment necessarily includes the particular features, structures, or characteristics. Further, some embodiments may have some, all, or none of the features described for other embodiments.
In the following description and claims, the term “coupled” along with its derivatives, may be used. “Coupled” is used to indicate that two or more elements co-operate or interact with each other, but they may or may not have intervening physical or electrical components between them.
As used in the claims, unless otherwise specified the use of the ordinal adjectives “first”, “second”, “third”, etc., to describe a common element, merely indicate that different instances of like elements are being referred to, and are not intended to imply that the elements so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
The drawings and the forgoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, orders of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions any flow diagram need not be implemented in the order shown; nor do all of the acts necessarily need to be performed. Also, those acts that are not dependent on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples. Numerous variations, whether explicitly given in the specification or not, such as differences in structure, dimension, and use of material, are possible. The scope of embodiments is at least as broad as given by the following claims.
The techniques shown in the figures can be implemented using code and data stored and executed on one or more electronic devices (e.g., an end station, a network element). Such electronic devices store and communicate (internally and/or with other electronic devices over a network) code and data using computer-readable media, such as non-transitory computer-readable storage media (e.g., magnetic disks; optical disks; random access memory; read only memory; flash memory devices; phase-change memory) and transitory computer-readable transmission media (e.g., electrical, optical, acoustical or other form of propagated signals—such as carrier waves, infrared signals, digital signals). In addition, such electronic devices typically include a set of one or more processors coupled to one or more other components, such as one or more storage devices (non-transitory machine-readable storage media), user input/output devices (e.g., a keyboard, a touchscreen, and/or a display), and network connections. The coupling of the set of processors and other components is typically through one or more busses and bridges (also termed as bus controllers). Thus, the storage device of a given electronic device typically stores code and/or data for execution on the set of one or more processors of that electronic device. Of course, one or more parts of an embodiment of the invention may be implemented using different combinations of software, firmware, and/or hardware.
In the foregoing specification, the invention has been described with reference to specific exemplary embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims. The Specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/US2011/055484 | 10/7/2011 | WO | 00 | 6/29/2013 |
