CPU time
CPU time is the amount of processor execution time consumed by a program, process, thread, or computation.
Experimentation for Engineers
CPU time is the amount of processor execution time consumed by a program or process.
- How it is used
- Uses CPU time as the metric minimized while tuning JIT parameters for web-server request handling.
- How deep it goes
- Works through noisy offline measurement, response-surface estimation, Gaussian process regression, and an iterative stopping rule.
- Examples
- Uses production request-log replay, a Python JIT-plus-server simulator, and measurements at parameters 0.5 and 0.0.
- What you can do after
- Apply it via Bayesian optimization to compare JIT configurations and stop when CPU-time measurements stabilize.
Build High Value Software Systems
The time spent actively progressing a change through activities such as coding, testing, fixing, and reviewing.
- How it is used
- Separates active progress from waiting so delivery delays can be investigated rather than attributed to slow coding.
- How deep it goes
- Defines what counts, distinguishes queues from active steps, and works through ratios and before-and-after station measurements.
- Examples
- Uses a 288-unit workflow with 40 active units and a review station whose elapsed time falls while active work stays constant.
- What you can do after
- Practice measuring active time on completed changes, locating waiting, and checking whether a targeted intervention improves flow.
Read how Build High Value Software Systems explains CPU time →
Troubleshooting Java: Read, debug, and optimize JVM applications
The time a program spends executing instructions on the CPU.
- How it is used
- Distinguishes CPU work from waiting when investigating application latency and thread locks.
- How deep it goes
- Explains how to interpret sampled CPU and execution times, then directs unresolved waiting cases to detailed profiling.
- Examples
- Compares a five-second HTTP response wait with a method showing 700 ms CPU self time and 4,903 ms execution self time.
- What you can do after
- Diagnose/debug latency by comparing CPU and execution times and using lock profiling to investigate suspected blocking.
Read how Troubleshooting Java: Read, debug, and optimize JVM applications explains CPU time →
Troubleshooting Java, Second Edition
CPU time is the amount of processor time a method or operation actively spends executing instructions.
- How it is used
- Uses CPU time with execution time to investigate latency and distinguish computation from waiting.
- How deep it goes
- Explains the metric through profiler observations and contrasts it with execution time, stopping short of measurement internals.
- Examples
- Uses a five-second /demo HTTP delay where getResponseCode() spends execution time waiting with zero CPU time.
- What you can do after
- Diagnose latency by comparing CPU time with execution time in profiler stacks.
Read how Troubleshooting Java, Second Edition explains CPU time →
Automating API Delivery
Task time, or touch time, is the time during which work is actively performed on a process task, excluding waiting and interruptions.
- How it is used
- Tracks active effort at process steps during process improvement, alongside lead time and output-quality measures.
- How deep it goes
- Defines included and excluded time, distinguishes task time from lead time, and shows how to record it in a process map.
- Examples
- Uses Acme’s API design and review process and Jira tickets awaiting processing as examples.
- What you can do after
- Evaluate process steps by recording task time and lead time in a process map and time ladder.
Apache Kafka in Action
CPU time is the amount of processor execution time consumed by a client's requests on a broker.
- How it is used
- Protects brokers from client-driven overload by limiting request-processing capacity.
- How deep it goes
- Defines broker and thread-based enforcement, explains limitations, and demonstrates default, client-specific, and user-specific quotas.
- Examples
- Uses online-shop defaults, a newsletter producer restriction, a warehouse consumer allowance, and a network/I/O thread formula.
- What you can do after
- Operate/configure Kafka quotas that reserve request capacity and reduce overload risk for selected clients or users.
Also covered in
- Starting Data Analytics with Generative AI and Python
- Kafka Streams in Action, Second Edition
- Latency
- Delivering Machine Learning Projects: From design to deployment
- MongoDB 8.0 in Action, Third Edition
What to know first
- experimentationExperimentation is the planned investigation of a question or hypothesis by conducting trials or interventions and observing their outcomes.
- ChatGPTChatGPT is OpenAI’s conversational AI service, which lets users interact with generative AI models and related tools through natural-language prompts.
- CPUA central processing unit (CPU) is a computer processor that executes program instructions and performs general-purpose computational operations.
- throughputThroughput is the rate at which a system processes or completes work, data, or requests over a specified period.
Related concepts
- activity ratioused by · An activity ratio is the proportion of total lead time spent on active work rather than waiting or other non-progressing periods.
- overall execution timecontrasts with · Overall execution time is the elapsed duration of a specified program, operation, test, workload, or thread’s execution.
- timestampkind of · A timestamp is a value representing a specific point in time, commonly associated with an event or data item.
- meancontrasts with · The mean is a measure of central location equal to the arithmetic average of numerical observations, or the probability-weighted average of a random variable’s possible values.
- experimentationmeasured by · Experimentation is the planned investigation of a question or hypothesis by conducting trials or interventions and observing their outcomes.
- lead timecontrasts with · Lead time is the elapsed time from when work is initiated or becomes available until its result is completed, delivered, or made available to its recipient.
Questions about CPU time
What is CPU time?
CPU time is the amount of processor execution time consumed by a program, process, thread, or computation.
Which Manning books explain CPU time?
Experimentation for Engineers, Build High Value Software Systems, Troubleshooting Java: Read, debug, and optimize JVM applications, Troubleshooting Java, Second Edition, Automating API Delivery and Apache Kafka in Action.
What should I know before learning CPU time?
Manning books that teach CPU time usually introduce experimentation, ChatGPT, CPU and throughput first.
What is the difference between CPU time and overall execution time?
CPU time is the amount of processor execution time consumed by a program, process, thread, or computation. Overall execution time is the elapsed duration of a specified program, operation, test, workload, or thread’s execution.
What is CPU time a kind of?
CPU time is a kind of timestamp. A timestamp is a value representing a specific point in time, commonly associated with an event or data item.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.





