Welcome!

Related Topics: @CloudExpo, Java IoT, Microservices Expo, Containers Expo Blog, @BigDataExpo, SDN Journal

@CloudExpo: Book Excerpt

Book Excerpt: Systems Performance: Enterprise and the Cloud | Part 1

CPUs drive all software and are often the first target for systems performance analysis

"This excerpt is from the book, "Systems Performance: Enterprise and the Cloud", authored by Brendan Gregg, published by Prentice Hall Professional, Oct. 2013, ISBN 9780133390094, Copyright © 2014 Pearson Education, Inc. For more info, please visit the publisher site:

CPUs drive all software and are often the first target for systems performance analysis. Modern systems typically have many CPUs, which are shared among all running software by the kernel scheduler. When there is more demand for CPU resources than there are resources available, process threads (or tasks) will queue, waiting their turn. Waiting can add significant latency during the runtime of applications, degrading performance.

The usage of the CPUs can be examined in detail to look for performance improvements, including eliminating unnecessary work. At a high level, CPU usage by process, thread, or task can be examined. At a lower level, the code path within applications and the kernel can be profiled and studied. At the lowest level, CPU instruction execution and cycle behavior can be studied.

This chapter consists of five parts:

  • Background introduces CPU-related terminology, basic models of CPUs, and key CPU performance concepts.
  • Architecture introduces processor and kernel scheduler architecture.
  • Methodology describes performance analysis methodologies, both observa- tional and experimental.
  • Analysis describes CPU performance analysis tools on Linux- and Solaris- based systems, including profiling, tracing, and visualizations.
  • Tuning includes examples of tunable parameters.

The first three sections provide the basis for CPU analysis, and the last two show its practical application to Linux- and Solaris-based systems.

The effects of memory I/O on CPU performance are covered, including CPU cycles stalled on memory and the performance of CPU caches. Chapter 7, Memory, continues the discussion of memory I/O, including MMU, NUMA/UMA, system interconnects, and memory busses.

Terminology
For reference, CPU-related terminology used in this chapter includes the following:

  • Processor: the physical chip that plugs into a socket on the system or pro- cessor board and contains one or more CPUs implemented as cores or hard- ware threads.
  • Core: an independent CPU instance on a multicore processor. The use of cores is a way to scale processors, called chip-level multiprocessing (CMP).
  • Hardware thread: a CPU architecture that supports executing multiple threads in parallel on a single core (including Intel's Hyper-Threading Tech- nology), where each thread is an independent CPU instance. One name for this scaling approach is multithreading.
  • CPU instruction: a single CPU operation, from its instruction set. There are instructions for arithmetic operations, memory I/O, and control logic.
  • Logical CPU: also called a virtual processor,1 an operating system CPU instance (a schedulable CPU entity). This may be implemented by the processor as a hardware thread (in which case it may also be called a virtual core), a core, or a single-core processor.
  • Scheduler: the kernel subsystem that assigns threads to run on CPUs.
  • Run queue: a queue of runnable threads that are waiting to be serviced by
  • CPUs. For Solaris, it is often called a dispatcher queue.

Other terms are introduced throughout this chapter. The Glossary includes basic terminology for reference, including CPU, CPU cycle, and stack. Also see the terminology sections in Chapters 2 and 3.

Models
The following simple models illustrate some basic principles of CPUs and CPU per- formance. Section 6.4, Architecture, digs much deeper and includes implementation- specific details.

CPU Architecture
Figure 1 shows an example CPU architecture, for a single processor with four cores and eight hardware threads in total. The physical architecture is pictured, along with how it is seen by the operating system.

Figure 1: CPU architecture

Each hardware thread is addressable as a logical CPU, so this processor appears as eight CPUs. The operating system may have some additional knowledge of topology, such as which CPUs are on the same core, to improve its scheduling decisions.

CPU Memory Caches
Processors provide various hardware caches for improving memory I/O perfor- mance. Figure 2 shows the relationship of cache sizes, which become smaller and faster (a trade-off) the closer they are to the CPU.

The caches that are present, and whether they are on the processor (integrated) or external to the processor, depend on the processor type. Earlier processors pro- vided fewer levels of integrated cache.

Figure 2: CPU cache sizes

CPU Run Queues
Figure 3 shows a CPU run queue, which is managed by the kernel scheduler.

Figure 3: CPU run queue

The thread states shown in the figure, ready to run and on-CPU, are covered in Figure 3.7 in Chapter 3, Operating Systems.

The number of software threads that are queued and ready to run is an impor- tant performance metric indicating CPU saturation. In this figure (at this instant) there are four, with an additional thread running on-CPU. The time spent waiting on a CPU run queue is sometimes called run-queue latency or dispatcher-queue latency. In this book, the term scheduler latency is used instead, as it is appropri- ate for all dispatcher types, including those that do not use queues (see the discus- sion of CFS in Section 6.4.2, Software).

For multiprocessor systems, the kernel typically provides a run queue for each CPU and aims to keep threads on the same run queue. This means that threads are more likely to keep running on the same CPUs, where the CPU caches have cached their data. (These caches are described as having cache warmth, and the approach to favor CPUs is called CPU affinity.) On NUMA systems, memory locality may also be improved, which also improves performance (this is described in Chapter 7, Memory).

It also avoids the cost of thread synchronization (mutex locks) for queue operations, which would hurt scalability if the run queue was global and shared among all CPUs.

Concepts
The following are a selection of important concepts regarding CPU performance, beginning with a summary of processor internals: the CPU clock rate and how instructions are executed. This is background for later performance analysis, particularly for understanding the cycles-per-instruction (CPI) metric.

Clock Rate
The clock is a digital signal that drives all processor logic. Each CPU instruction may take one or more cycles of the clock (called CPU cycles) to execute. CPUs exe- cute at a particular clock rate; for example, a 5 GHz CPU performs 5 billion clock cycles per second.

Some processors are able to vary their clock rate, increasing it to improve performance or decreasing it to reduce power consumption. The rate may be varied on request by the operating system, or dynamically by the processor itself. The ker- nel idle thread, for example, can request the CPU to throttle down to save power.

Clock rate is often marketed as the primary feature of the processor, but this can be a little misleading. Even if the CPU in your system appears to be fully utilized (a bottleneck), a faster clock rate may not speed up performance-it depends on what those fast CPU cycles are actually doing. If they are mostly stall cycles while waiting on memory access, executing them more quickly doesn't actually increase the CPU instruction rate or workload throughput.

Instruction
CPUs execute instructions chosen from their instruction set. An instruction includes the following steps, each processed by a component of the CPU called a functional unit:

  1. Instruction fetch
  2. Instruction decode
  3. Execute
  4. Memory access
  5. Register write-back

The last two steps are optional, depending on the instruction. Many instructions operate on registers only and do not require the memory access step.

Each of these steps takes at least a single clock cycle to be executed. Memory access is often the slowest, as it may take dozens of clock cycles to read or write to main memory, during which instruction execution has stalled (and these cycles while stalled are called stall cycles). This is why CPU caching is important, as described in Section 6.4: it can dramatically reduce the number of cycles needed for memory access.

Instruction Pipeline
The instruction pipeline is a CPU architecture that can execute multiple instructions in parallel, by executing different components of different instructions at the same time. It is similar to a factory assembly line, where stages of production can be executed in parallel, increasing throughput.

Consider the instruction steps previously listed. If each were to take a single clock cycle, it would take five cycles to complete the instruction. At each step of this instruction, only one functional unit is active and four are idle. By use of pipe- lining, multiple functional units can be active at the same time, processing differ- ent instructions in the pipeline. Ideally, the processor can then complete one instruction with every clock cycle.

Instruction Width
But we can go faster still. Multiple functional units can be included of the same type, so that even more instructions can make forward progress with each clock cycle. This CPU architecture is called superscalar and is typically used with pipe- lining to achieve a high instruction throughput.

The instruction width describes the target number of instructions to process in parallel. Modern processors are 3-wide or 4-wide, meaning they can complete up to three or four instructions per cycle. How this works depends on the processor, as there may be different numbers of functional units for each stage.

CPI, IPC
Cycles per instruction (CPI) is an important high-level metric for describing where a CPU is spending its clock cycles and for understanding the nature of CPU utilization. This metric may also be expressed as instructions per cycle (IPC), the inverse of CPI.

A high CPI indicates that CPUs are often stalled, typically for memory access. A low CPI indicates that CPUs are often not stalled and have a high instruction throughput. These metrics suggest where performance tuning efforts may be best spent.

Memory-intensive workloads, for example, may be improved by installing faster memory (DRAM), improving memory locality (software configuration), or reducing the amount of memory I/O. Installing CPUs with a higher clock rate may not improve performance to the degree expected, as the CPUs may need to wait the same amount of time for memory I/O to complete. Put differently, a faster CPU may mean more stall cycles but the same rate of completed instructions.

The actual values for high or low CPI are dependent on the processor and processor features and can be determined experimentally by running known work- loads. As an example, you may find that high-CPI workloads run with a CPI at ten or higher, and low CPI workloads run with a CPI at less than one (which is possi- ble due to instruction pipelining and width, described earlier).

It should be noted that CPI shows the efficiency of instruction processing, but not of the instructions themselves. Consider a software change that added an inefficient software loop, which operates mostly on CPU registers (no stall cycles): such a change may result in a lower overall CPI, but higher CPU usage and utilization.

Utilization
CPU utilization is measured by the time a CPU instance is busy performing work during an interval, expressed as a percentage. It can be measured as the time a CPU is not running the kernel idle thread but is instead running user-level application threads or other kernel threads, or processing interrupts.

High CPU utilization may not necessarily be a problem, but rather a sign that the system is doing work. Some people also consider this an ROI indicator: a highly utilized system is considered to have good ROI, whereas an idle system is considered wasted. Unlike with other resource types (disks), performance does not degrade steeply under high utilization, as the kernel supports priorities, preemption, and time sharing. These together allow the kernel to understand what has higher priority, and to ensure that it runs first.

The measure of CPU utilization spans all clock cycles for eligible activities, including memory stall cycles. It may seem a little counterintuitive, but a CPU may be highly utilized because it is often stalled waiting for memory I/O, not just executing instructions, as described in the previous section.

CPU utilization is often split into separate kernel- and user-time metrics.

More Stories By Brendan Gregg

Brendan Gregg, Lead Performance Engineer at Joyent, analyzes performance and scalability throughout the software stack. As Performance Lead and Kernel Engineer at Sun Microsystems (and later Oracle), his work included developing the ZFS L2ARC, a pioneering file system technology for improving performance using flash memory. He has invented and developed many performance tools, including some that ship with Mac OS X and Oracle® Solaris™ 11. His recent work has included performance visualizations for Linux and illumos kernel analysis. He is coauthor of DTrace (Prentice Hall, 2011) and Solaris™ Performance and Tools (Prentice Hall, 2007).

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


Latest Stories
Basho Technologies has announced the latest release of Basho Riak TS, version 1.3. Riak TS is an enterprise-grade NoSQL database optimized for Internet of Things (IoT). The open source version enables developers to download the software for free and use it in production as well as make contributions to the code and develop applications around Riak TS. Enhancements to Riak TS make it quick, easy and cost-effective to spin up an instance to test new ideas and build IoT applications. In addition to...
You think you know what’s in your data. But do you? Most organizations are now aware of the business intelligence represented by their data. Data science stands to take this to a level you never thought of – literally. The techniques of data science, when used with the capabilities of Big Data technologies, can make connections you had not yet imagined, helping you discover new insights and ask new questions of your data. In his session at @ThingsExpo, Sarbjit Sarkaria, data science team lead ...
Many private cloud projects were built to deliver self-service access to development and test resources. While those clouds delivered faster access to resources, they lacked visibility, control and security needed for production deployments. In their session at 18th Cloud Expo, Steve Anderson, Product Manager at BMC Software, and Rick Lefort, Principal Technical Marketing Consultant at BMC Software, discussed how a cloud designed for production operations not only helps accelerate developer in...
Extracting business value from Internet of Things (IoT) data doesn’t happen overnight. There are several requirements that must be satisfied, including IoT device enablement, data analysis, real-time detection of complex events and automated orchestration of actions. Unfortunately, too many companies fall short in achieving their business goals by implementing incomplete solutions or not focusing on tangible use cases. In his general session at @ThingsExpo, Dave McCarthy, Director of Products...
Ask someone to architect an Internet of Things (IoT) solution and you are guaranteed to see a reference to the cloud. This would lead you to believe that IoT requires the cloud to exist. However, there are many IoT use cases where the cloud is not feasible or desirable. In his session at @ThingsExpo, Dave McCarthy, Director of Products at Bsquare Corporation, will discuss the strategies that exist to extend intelligence directly to IoT devices and sensors, freeing them from the constraints of ...
WebRTC is bringing significant change to the communications landscape that will bridge the worlds of web and telephony, making the Internet the new standard for communications. Cloud9 took the road less traveled and used WebRTC to create a downloadable enterprise-grade communications platform that is changing the communication dynamic in the financial sector. In his session at @ThingsExpo, Leo Papadopoulos, CTO of Cloud9, discussed the importance of WebRTC and how it enables companies to focus...
Venafi has extended the power of its platform in an easy-to-use utility for DevOps teams available for immediate download. Now DevOps teams can eliminate the hassle of acquiring and installing TLS keys and certificates. Instead, customers can focus on speeding up continuous development and deployment, while security teams have complete visibility and can keep the DevOps environment secure and compliant to protect customer data. Extending the Venafi Trust Protection Platform requires only a singl...
Aspose.Total for .NET is the most complete package of all file format APIs for .NET as offered by Aspose. It empowers developers to create, edit, render, print and convert between a wide range of popular document formats within any .NET, C#, ASP.NET and VB.NET applications. Aspose compiles all .NET APIs on a daily basis to ensure that it contains the most up to date versions of each of Aspose .NET APIs. If a new .NET API or a new version of existing APIs is released during the subscription peri...
The competitive landscape of the global cloud computing market in the healthcare industry is crowded due to the presence of a large number of players. The large number of participants has led to the fragmented nature of the market. Some of the major players operating in the global cloud computing market in the healthcare industry are Cisco Systems Inc., Carestream Health Inc., Carecloud Corp., AGFA Healthcare, IBM Corp., Cleardata Networks, Merge Healthcare Inc., Microsoft Corp., Intel Corp., an...
The best-practices for building IoT applications with Go Code that attendees can use to build their own IoT applications. In his session at @ThingsExpo, Indraneel Mitra, Senior Solutions Architect & Technology Evangelist at Cognizant, provided valuable information and resources for both novice and experienced developers on how to get started with IoT and Golang in a day. He also provided information on how to use Intel Arduino Kit, Go Robotics API and AWS IoT stack to build an application tha...
Is your aging software platform suffering from technical debt while the market changes and demands new solutions at a faster clip? It’s a bold move, but you might consider walking away from your core platform and starting fresh. ReadyTalk did exactly that. In his General Session at 19th Cloud Expo, Michael Chambliss, Head of Engineering at ReadyTalk, will discuss why and how ReadyTalk diverted from healthy revenue and over a decade of audio conferencing product development to start an innovati...
With an estimated 50 billion devices connected to the Internet by 2020, several industries will begin to expand their capabilities for retaining end point data at the edge to better utilize the range of data types and sheer volume of M2M data generated by the Internet of Things. In his session at @ThingsExpo, Don DeLoach, CEO and President of Infobright, discussed the infrastructures businesses will need to implement to handle this explosion of data by providing specific use cases for filterin...
Cloud analytics is dramatically altering business intelligence. Some businesses will capitalize on these promising new technologies and gain key insights that’ll help them gain competitive advantage. And others won’t. Whether you’re a business leader, an IT manager, or an analyst, we want to help you and the people you need to influence with a free copy of “Cloud Analytics for Dummies,” the essential guide to this explosive new space for business intelligence.
So, you bought into the current machine learning craze and went on to collect millions/billions of records from this promising new data source. Now, what do you do with them? Too often, the abundance of data quickly turns into an abundance of problems. How do you extract that "magic essence" from your data without falling into the common pitfalls? In her session at @ThingsExpo, Natalia Ponomareva, Software Engineer at Google, provided tips on how to be successful in large scale machine learning...
Early adopters of IoT viewed it mainly as a different term for machine-to-machine connectivity or M2M. This is understandable since a prerequisite for any IoT solution is the ability to collect and aggregate device data, which is most often presented in a dashboard. The problem is that viewing data in a dashboard requires a human to interpret the results and take manual action, which doesn’t scale to the needs of IoT.