Patent Application
20150020076
The present application relates generally to computers, and computer applications, and more particularly to bottleneck detection and capacity planning in computing systems.
Software performance depends on allocated resources. For instance, when software runs in a computer system, whether it is a centralized or a distributed system, the software performance depends on how much of computing resources is being allocated. Computer resources may include physical resources as well as virtual resources. Examples of physical resources, e.g., may include a processor (central processing unit (CPU) or CPU cycles), memory, internal Input/Output (disk I/O) and external I/O (networking bandwidth). Examples of virtual resources may include thread pool size, buffer size and others.
Existing approaches that try to understand a component's performance may require experts with deep understanding on the software structure, i.e., how to break the software into individual components, and how the end-to-end performance is derived from the component-level performance. Such approaches may also require adding instrumenting code to the software source code which may be difficult in many situations. Further such approaches may be highly software-specific, and cannot be easily generalized.
Another known approach may intercept the targeted software. The drawback of this approach lies in the difficulty for designing the intercept. It also requires the knowledge of the APIs for all the low-level resource usage that the software is calling, which presents a difficult practice. In addition, the large overhead created by such monitoring layer or intercept may significantly affect the software performance such that the collected performance numbers are not representative. Other approaches may passively collect observed data for performance related information, but with such approaches the effectiveness may highly depend on the observed data only.
A method of detecting resource bottleneck of a software entity, in one aspect, may comprise identifying a resource type in a system executing the software entity. The method may also comprise intentionally inducing perturbation by varying the supply amount of the resource type in the system. The method may further comprise measuring performance of the software entity at multiple variation levels of the supply amount of the resource type in the system. The method may also comprise modeling a relationship between the measured performance and the variation levels. The method may further comprise applying the model to detect the resource bottleneck.
A system for detecting resource bottleneck of a software entity, in one aspect, may comprise a module operable to execute on the processor and further operable to identify a resource type in a system executing the software entity, the module further operable to intentionally induce perturbation by varying the supply amount of the resource type in the system. The module may be further operable to measure performance of the software entity at multiple variation levels of the supply amount of the resource type in the system. The module may be further operable to model a relationship between the measured performance and the variation levels. The system may also comprise a bottleneck detector operable to apply the model to detect the resource bottleneck.
A computer readable storage medium storing a program of instructions executable by a machine to perform one or more methods described herein also may be provided.
Further features as well as the structure and operation of various embodiments are described in detail below with reference to the accompanying drawings. In the drawings, like reference numbers indicate identical or functionally similar elements.
Detecting resource bottleneck and planning capacity are two closely relevant issues and can usually be solved simultaneously. Bottleneck detection refers to identifying the resource type(s) that restrains or prevents the system performance from improving, even with an increase of the other types of resources. The bottleneck resources could be a single type or a combination of multiple types. Capacity planning refers to determining how much performance gain should be expected, if an extra amount of bottleneck resources that combat bottleneck are added. For instance, a frequently occurring problem is how to detect which resource is the bottleneck that limits the software performance. Another relevant problem is, if one resource type is detected as the bottleneck and an extra amount of the resource is added to combat the detected bottleneck, determining how much performance gain can be expected. Such a two-stage problem in the present disclosure is called resource bottleneck detection and capacity planning. It is a general problem, applied to not only software applications, but also operating system, virtual machines and others. Bottleneck detection and capacity planning are useful in a wide range of scenarios. For instance, bottleneck detection may be needed for debugging software performance and scaling up. Capacity planning may be often needed for estimating the Return-of-investment (ROI) or cloud resource.
In one embodiment of the present disclosure, resource bottleneck detection and capacity planning is provided. The resource bottleneck detection and capacity planning of the present disclosure in one embodiment is general enough to be applicable to many scenarios and applications. The effectiveness of resource bottleneck detection and capacity planning of the present disclosure in one embodiment does not depend on data that are passively observed.
A methodology in one embodiment of the present disclosure may induce positive or negative perturbation to a resource supply by imposing various or varying perturbation levels. In this way, one can observe how an end-to-end performance changes based on the varying perturbation levels. The methodology of the present disclosure in one embodiment may model the relationship between the performance measure and the perturbation level. The model can be used to predict performance improvement/degradation whenever the resource supply is increased or decreased.
For example, targeted software uses a portion of a system resource and its performance is correlated to the resource supply. The resource supply may be perturbed and how the targeted software's performance reacts may be modeled. The derived model reflects how the perturbation impacts the performance. By carefully designing the perturbation experiment, specifically, selecting the resource types that are perturbed, and determining the amount of resources changed by the perturbation, the methodology of the present disclosure in one embodiment may compare and find out which resource has the most significant impact on the software performance; such a resource could be identified as the bottleneck. There could be more than one resource identified as the bottlenecks.
The methodology of the present disclosure in one embodiment need not require expert knowledge on the system or software, or another whose performance is impacted, and hence may be easy to adapt. In addition, the methodology of the present disclosure in one embodiment is very general in that the method may be applicable to a broad spectrum of scenarios, including e.g., software on a single system, distributed software on a distributed systems, virtual machines deployed on physical machines. The perturbation described in the present embodiment may perturb the aggregate resource supply, that is, the resource provided for the entire system instead of for individual software components. This makes the perturbation very practical. For instance, if the resource type is a CPU, the perturbation can be conducted by varying the number of enabled CPU cores before the system bootup, e.g., rather than controlling the CPU cycles allocated to individual running processes. As another example, if the concerned resource is memory size, the aggregate resource supply is the total physical memory available for the OS (Operating System). In this case, perturbing the aggregate resource supply can be realized by adding a kernel boot parameter to specify the maximum available physical memory (for Linux environments). The methodology of the present disclosure in one embodiment is not limited to those fundamental physical resources such as CPU, I/O, network bandwidth and memory. Rather, the methodology of the present disclosure in one embodiment is applicable to any virtual or scenario-specific resources, e.g., virtual memory size, thread pool size, and others. In one aspect, the methodology of the present disclosure utilizes a way to control the supplied resource amount.
For example, the methodology of the present disclosure in one embodiment may emulate a slower link by adding a packet delay to each outgoing packet on the database server 106. The introduced extra delay is the perturbation in this example. The methodology of the present disclosure in one embodiment may then measure the application performance by running the DTW for a sufficient period of time. By testing a set of extra delays, the methodology of the present disclosure in one embodiment may model how the performance (e.g., request response time) changes with respect to the extra packet delay.
The measured data points may be used to estimate how much the response time will reduce if the 1 Gbps Ethernet link is replaced by a faster fabric. Because replacing the link by a faster fabric is equivalent to injecting a negative extra packet delay, the relationship shown in
Referring to
At 404, for each resource type, the aggregate resource supply for that resource is varied by a determined delta amount. That is, variation is induced in the supply amount of the resource. The delta amount or variation gradient may be determined based on the characteristic or property of the resource and/or software or component whose performance is being measured.
Introducing perturbation to a system resource supply may be realized in different ways. In one embodiment, the different ways of inducing perturbation to system resources may depend on the platform and resource type. The following examples illustrate a few scenarios, however, the methodology of the present disclosure is not limited only to those examples. For example, perturbation may be introduced in CPU related resources, for instance, by using CPU frequency scaling to reduce the CPU processing speed. If a multi-core CPU is used and the resource in concern is CPU cores, perturbation may be introduced by adding a parameter in the OS booting configuration to disable one or more cores. As another example, perturbation may be induced in memory related resources. For instance, the physical memory size for a computer can be made partially usable in operating system (OS) booting configurations. As yet another example, networking I/O related resources may be perturbed or varied. For instance, available network bandwidth can be controlled by traffic rate controller, which usually can be found in a quality of service (QoS) module of most modern operating systems. Network latency can be purposely increased by tools such as netem in LINUX. Still yet as another example, disk I/O related resource may be perturbed or varied. For instance, on UNIX-like systems, disk I/O may be perturbed by tools such as ionice, which sets the I/O scheduling class and priority for a program or script. One can also use rsync with an option “--bwlimit” to control I/O bandwidth. Other systems may provide similar tools or commands for varying disk I/O. Other resources that can be perturbed or varied may include, but are not limited to, any configurable resources, e.g., virtual resources such as thread pool size and file descriptors. Perturbation experiment can be performed by changing such parameters. In one aspect, the software or a component whose performance is being measured may be restarted for measuring the performance. In one aspect, when varying the total resource supply, finer-granularity variation may be applied when the variation reaches close to zero unit.
At 406, the performance is measured. The processing returns to 404, where the perturbation is varied again, and at 406 the performance is measured again with the variance in the perturbation. The processing at 404 and 406, such that performance is measured at multiple variation levels of the supply associated with the resource.
At 408, a model is built. The collected set of points (performance measure at multiple variation levels) provides the performance measure for various levels of resource supply. Based on the measure, the method may estimate the performance for a certain level of resource supply that has not been or is impossible to be experimented. In one embodiment, this may be formulated as the classical model interpolation problem: based on the output data from various experimental inputs, the output for similar input that has not been tested is predicted. Because the measured data may be subject to noise, an appropriate method may be selected and applied to solve this data interpolation problem. The interpolated model, e.g., may include parameters that allow one to compute the output for any given input. The quality of the interpolated model can be measured by various goodness of fit statistical testing such as chi-square test. In addition to classical interpolation methods such as linear regression, polynomial fitting and spline, more sophisticated machine learning techniques such as Neural Networks and Support Vector Machine can be exploited for the model interpolation. Based on the collected data, any one or more of the above techniques may be selected for building a model of the present disclosure in one embodiment. In one aspect, the method of the present disclosure may balance over-fitting with accuracy.
In one embodiment, the method of the present disclosure may apply perturbing to more than one resource simultaneously. In the embodiment where there are multiple resource types involved, the method of the present disclosure may solve a multi-dimensional model type interpolation problem. The following example is a two-dimensional model. For example, referring to the DTW scenario described with respect to
Performance=f(CPU usage, packet latency),
where f(.) is an empirically fitted function. Such a model is a two-dimensional model since f(.) contains two metrics. When more resource types are involved, it can be extended to a multiple-dimensional model.
The method may perform the processing at 404, 406 and 408 for more than one resource type. Hence, the method may repeat 404, 406 and 408.
At 410, it is determined whether the built model is considered statistically sound by statistical criteria such as chi-square test. If so, the model may be utilized. If not, the processing may return to 404 where another model may be built, e.g., based on different variation levels.
At 412, the built model (e.g., via model interpolation) may be applied to determine capacity planning. Capacity planning refers to predicting performance changes when the resource supply varies. Capacity planning can also refer to determining the amount of needed resources if certain Service-level-agreement (SLA) needs to be achieved. The model derived earlier can be used for the capacity planning purpose, either for single resource or multiple resources.
At 414, the method of the present disclosure may further detect which resource is the performance bottleneck, i.e., by perturbing resources and creating the model interpolation for each individual resource type. By comparing how significant the performance is affected by a slight variation of each individual resource's supply, the method of the present disclosure may determine which resource is the bottleneck. For instance, in the above DTW example scenario, one can first perturb the network latency and observe how the performance changes. If the performance is quite stable even though the network latency increased a lot, it indicates that the network latency is not the bottleneck. Consequently, one can perturb the CPU usage by using methods such as CPU frequency scaling, if the performance dramatically changes, the CPU is most likely the bottleneck; otherwise, the perturbation test should further extend to the other resource types.
In one embodiment of the present disclosure, a user interface such as a graphical user interface may be provided for providing to a user one or more outputs from the above described methodologies, e.g., the modeled relationship and data analysis.
The computer system may be described in the general context of computer system executable instructions, such as program modules, being executed by a computer system. Generally, program modules may include routines, programs, objects, components, logic, data structures, and so on that perform particular tasks or implement particular abstract data types. The computer system may be practiced in distributed cloud computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules may be located in both local and remote computer system storage media including memory storage devices.
The components of computer system may include, but are not limited to, one or more processors or processing units 12, a system memory 16, and a bus 14 that couples various system components including system memory 16 to processor 12. The processor 12 may include a module 10 that performs the methods described herein. The module 10 may be programmed into the integrated circuits of the processor 12, or loaded from memory 16, storage device 18, or network 24 or combinations thereof.
Bus 14 may represent one or more of any of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MCA) bus, Enhanced ISA (EISA) bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnects (PCI) bus.
Computer system may include a variety of computer system readable media. Such media may be any available media that is accessible by computer system, and it may include both volatile and non-volatile media, removable and non-removable media.
System memory 16 can include computer system readable media in the form of volatile memory, such as random access memory (RAM) and/or cache memory or others. Computer system may further include other removable/non-removable, volatile/non-volatile computer system storage media. By way of example only, storage system 18 can be provided for reading from and writing to a non-removable, non-volatile magnetic media (e.g., a “hard drive”). Although not shown, a magnetic disk drive for reading from and writing to a removable, non-volatile magnetic disk (e.g., a “floppy disk”), and an optical disk drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM or other optical media can be provided. In such instances, each can be connected to bus 14 by one or more data media interfaces.
Computer system may also communicate with one or more external devices 26 such as a keyboard, a pointing device, a display 28, etc.; one or more devices that enable a user to interact with computer system; and/or any devices (e.g., network card, modem, etc.) that enable computer system to communicate with one or more other computing devices. Such communication can occur via Input/Output (I/O) interfaces 20.
Still yet, computer system can communicate with one or more networks 24 such as a local area network (LAN), a general wide area network (WAN), and/or a public network (e.g., the Internet) via network adapter 22. As depicted, network adapter 22 communicates with the other components of computer system via bus 14. It should be understood that although not shown, other hardware and/or software components could be used in conjunction with computer system. Examples include, but are not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data archival storage systems, etc.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or computer program product. Accordingly, aspects of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment (including firmware, resident software, micro-code, etc.) or an embodiment combining software and hardware aspects that may all generally be referred to herein as a “circuit,” “module” or “system.” Furthermore, aspects of the present invention may take the form of a computer program product embodied in one or more computer readable medium(s) having computer readable program code embodied thereon.
Any combination of one or more computer readable medium(s) may be utilized. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the “C” programming language or similar programming languages, a scripting language such as Perl, VBS or similar languages, and/or functional languages such as Lisp and ML and logic-oriented languages such as Prolog. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).
Aspects of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer readable medium that can direct a computer, other programmable data processing apparatus, or other devices to function in a particular manner, such that the instructions stored in the computer readable medium produce an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The computer program product may comprise all the respective features enabling the implementation of the methodology described herein, and which—when loaded in a computer system—is able to carry out the methods. Computer program, software program, program, or software, in the present context means any expression, in any language, code or notation, of a set of instructions intended to cause a system having an information processing capability to perform a particular function either directly or after either or both of the following: (a) conversion to another language, code or notation; and/or (b) reproduction in a different material form.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
The corresponding structures, materials, acts, and equivalents of all means or step plus function elements, if any, in the claims below are intended to include any structure, material, or act for performing the function in combination with other claimed elements as specifically claimed. The description of the present invention has been presented for purposes of illustration and description, but is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the invention. The embodiment was chosen and described in order to best explain the principles of the invention and the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.
Various aspects of the present disclosure may be embodied as a program, software, or computer instructions embodied in a computer or machine usable or readable medium, which causes the computer or machine to perform the steps of the method when executed on the computer, processor, and/or machine. A program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine to perform various functionalities and methods described in the present disclosure is also provided.
The system and method of the present disclosure may be implemented and run on a general-purpose computer or special-purpose computer system. The terms “computer system” and “computer network” as may be used in the present application may include a variety of combinations of fixed and/or portable computer hardware, software, peripherals, and storage devices. The computer system may include a plurality of individual components that are networked or otherwise linked to perform collaboratively, or may include one or more stand-alone components. The hardware and software components of the computer system of the present application may include and may be included within fixed and portable devices such as desktop, laptop, and/or server. A module may be a component of a device, software, program, or system that implements some “functionality”, which can be embodied as software, hardware, firmware, electronic circuitry, or etc.
The embodiments described above are illustrative examples and it should not be construed that the present invention is limited to these particular embodiments. Thus, various changes and modifications may be effected by one skilled in the art without departing from the spirit or scope of the invention as defined in the appended claims.
