Latest news

NEW: MARCH 2024 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Thread-local handshakes allows sampling which does not suffer from safepoint bias and has very low overhead"
• All the latest Java performance tips extracted in concise form
  • "Steady-state and limit load tests are quite different. Steady-state models your expected traffic including peaks. Limit increases the load until some resource is saturated - this identifies the resource that will bottleneck your system, and also whether the system can recover if the load then reduces"

FEBRUARY 2024 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Java 22 has one significant new feature, the Foreign Function and Memory API - finally released after more than 10 years in development"
• All the latest Java performance tips extracted in concise form
  • "Arena models the lifecycle of MemorySegments. They can be Global (lasts for the JVM lifetime), Automatic (managed by GC), Confined (bounded to a single thread and deallocated when closed or scope exits) or Shared (not bounded by threads but also deallocated when closed or scope exits)"

JANUARY 2024 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "For concurrent updates Hashtable.size() is strictly consistent, ConcurrentHashMap.size() is eventually consistent"
• All the latest Java performance tips extracted in concise form
  • "Because garbage collection costs across an entire organization is hugely significant and garbage collection cost tends to be dominated by memory requirements and pressure, memory profiling and tuning actually reduces costs across the board as reduced memory pressure translates directly to reduced garbage collection which translates directly to reduced CPU! A lot of high-performance Java dev have become experts at avoiding allocations altogether for exactly this reason"

DECEMBER 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Eventual consistency has become the more popular mechanism because of the lower latency of data access"
• All the latest Java performance tips extracted in concise form
  • "Selecting the right garbage collector is crucial for optimizing application performance: Serial GC for small applications with low memory footprint and low CPU requirements, but not suitable for low-latency applications due to longer GC pauses; Parallel GC (Throughput Collector) for maximizing application throughput by utilizing multiple threads for garbage collection but also not suitable for low-latency applications due to significant GC pauses; CMS (Concurrent Mark Sweep) designed to minimize pause times by working concurrently with application threads, has lower pause times but suffers from fragmentation ultimately giving long pauses; G1 (Garbage-First Collector) tries to provide a good balance between throughput and pause times, but doesn't give the lowest pause times; ZGC (Z Garbage Collector) and Shenandoah designed for low-latency applications, aiming to achieve pause times of less than 10ms, even on large heaps"

NOVEMBER 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "The full process data structure of the JVM is dramatically different depending on which GC algorithm is chosen"
• All the latest Java performance tips extracted in concise form
  • "Common cache invalidation algorithms: Time-based after a certain period of time; Version-based whenever version changes; Delta-based - compare the current version with the latest version and invalidate if the delta between the two exceeds a certain threshold; Least Recently Used (LRU) - the item not accessed for the longest time is evicted from the cache; Least Frequently Used (LFU) - the item accessed the least is evicted from the cache. Of course caches can combine multiple algorithms"

OCTOBER 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Most JVM tuning changes needed in production are garbage collection related changes"
• All the latest Java performance tips extracted in concise form
  • "Testing gotchas: testing once is jittery and un-representative, interpreted code behaves very differently from compiled JIT optimized code, System.currentTimeMillis() can go backwards (eg from NTP), JVM code and branch elimination can give unrepresentative performance for unrepresentative data and workloads, testing on different platforms can give very different (unrepresentative) results, power saving modes can give unrepresentative results, having other processes running at the same time can give unrepresentative results."

SEPTEMBER 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "I show ALL the performance changes in EVERY released version of the JDK since JDK 10"
• All the latest Java performance tips extracted in concise form
  • "Virtual threads currently (JDK21) implement cooperative scheduling. This isn't a limitation of virtual threads, it's a choice of the current implementation. Virtual threads can implement fully pre-emptive scheduling (and may in future)"

AUGUST 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "There is just one significant new performance feature - but it's huge! Virtual threads have finally arrived"
• All the latest Java performance tips extracted in concise form
  • "Virtual threads run on daemon (carrier/platform) threads in a ForkJoin pool. So be aware that if there are no non-daemon threads left, the daemon threads - and so all virtual threads - will be terminated"

JULY 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "the most common performance concerns are mundane - here's how to deal with them"
• All the latest Java performance tips extracted in concise form
  • "If running on bare metal, set the heap to whatever size the application needs, but less than available RAM; for a container set to less than 75% of container memory. Set using -Xmx or -XX:MaxRAMPercentage=75 - but note applications and 3rd party products using off-heap memory may need this lower"

JUNE 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Great for monitoring, generating alerts, debugging, root cause analysis, system resource analysis"
• All the latest Java performance tips extracted in concise form
  • "When scaling to achieve the same operations, small scale implementations may need different algorithms than the large scale implementations if you want optimal efficiency, especially when considering parallelization"

MAY 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "this is, unusually, a very controversial JEP; there is a strong preference by the community that it should NOT proceed"
• All the latest Java performance tips extracted in concise form
  • "If you have performance problems from having huge objects regularly collected, you need to tune the region sizes or try a different garbage collector"

APRIL 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "This change, if it proceeds, will stop a large swathe of troubleshooting and monitoring tools from working on JVMs (without having to first restart them with an option change)"
• All the latest Java performance tips extracted in concise form
  • "Allocation rates of around 10GB/s is the maximum that the majority of machines can sustain - regardless of core counts. And at that rate it dominates CPU. Four threads allocating is already enough to hit the highest per-thread allocation rate possible on many systems. The L3 cache is the limiting factor"

MARCH 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "The new 4GLs aren't new languages, they are 3GLs together with intelligent IDEs"
• All the latest Java performance tips extracted in concise form
  • "Strategies to prevent deadlocks: always acquire locks in a fixed order; avoid holding locks for extended periods; reduce the scope of the lock holding code block; use ReentrantLock and similar concurrent lock classes to manage locks, eg using tryLock() to acquire locks without blocking; use timeouts when acquiring locks to prevent threads from waiting indefinitely; use non-blocking algorithms and data structures to avoid deadlocks altogether"

FEBRUARY 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Java 20 large performance changes are all preview features; the small changes are GC and HTTP related"
• All the latest Java performance tips extracted in concise form
  • "Trivially copyable objects (where the majority or entire Java object can be copied as a memory copy without any serialization logic) provides optimal efficiency when moving object data across services"

JANUARY 2023 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Performance is naturally chaotic - you should really know and plan to apply chaos testing"
• All the latest Java performance tips extracted in concise form
  • "Metrics give you a numeric value at a point in time and are easily aggregated. They are great for telling you something has happened, but not always what, and not at all why. Logs tell you what, and let you determine why. Traces tell you where"

DECEMBER 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "CRaC can let you effectively create a Java binary running a full JVM that starts after all the app initialization and JIT compilation is done!"
• All the latest Java performance tips extracted in concise form
  • "Cold start optimization techniques: Class Data Sharing in OpenJDK, Shared class cache in OpenJ9, Ahead of time compilation eg Graal native image, OpenJDK CRaC (saved image), OpenJDK Leyden (static binaries), Alibaba Dragonwell JWarmUp, jlink/jpackage removed modules"

NOVEMBER 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "OpenTelemetry provides a fully implemented Java SDK and agent which allows your application to be monitored - for free! But the data needs to go somewhere, be stored, analyzed and presented - and you don't get that as part of OpenTelemetry"
• All the latest Java performance tips extracted in concise form
  • "Identify that a service cannot handle any more requests by monitoring for : increasing service response times, how long requests are queued for, and the count of dropped requests"

OCTOBER 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "These six pillars have different data types, all with different structures and strengths for identifying whether issues are occurring and why."
• All the latest Java performance tips extracted in concise form
  • "TEMPLE, the six pillars of observability: traces, events, metrics, profiles, logs, and exceptions"

SEPTEMBER 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "More than 10 different major organizations each providing their own freely available JVM, with support"
• All the latest Java performance tips extracted in concise form
  • "Use sampling profiling before bytecode profiling, as sampling has a much lower overhead and often gives you enough information to optimize - except for reactive apps"

AUGUST 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Virtual Threads has been released as a preview!"

JULY 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "DO optimize some things early - re-architecting your application is very expensive and effortful, especially when some thought at the beginning of the project let's you avoid having to make costly optimizations later"
• All the latest Java performance tips extracted in concise form
  • "A common mistake for micro performance optimizations is not preventing the system from having the same performance degradation in the future after maintenance changes. A complete fix would require regression tests that prevent the improvement from degrading by monitoring the performance"

JUNE 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Now with very little effort, you can create your own custom full featured Java agent!"
• All the latest Java performance tips extracted in concise form
  • "Request handling entry and exit points are often a good point to target instrumentation."

MAY 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "the configuration to ensure responsiveness needs to consider failure modes including failover and startup"
• All the latest Java performance tips extracted in concise form
  • "There are always queues somewhere. Queue size is NOT a useful metric, since a non-empty queue does not imply a problem, in fact it could imply an efficient use of resources. Focus instead on the metrics that affect response time - median task wait time. The longest wait times are often anomalous - or may need to be targeted specifically depending on your application"

APRIL 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "The techniques used to make it the lowest possible overhead profiler: infrequent sampling, no stack walking, no thread matching or analysis, minimal processing of data"
• All the latest Java performance tips extracted in concise form
  • "A cache pattern decision chart: 1 - if your data is crucial for startup and not highly dynamic, use a file persistent storage mechanism closer to the services to store the data; 2 - if your data is highly dynamic, use a dedicated caching technology that can handle this, with caches closer to the service or lower latency than the primary data store; 3 - if your data is not highly dynamic and can be easily invalidated, use a reverse proxy"

MARCH 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "I explain how to use OpenTelemetry, both with no code changes and by adding custom metrics"
• All the latest Java performance tips extracted in concise form
  • "Keeping a small working set of data improves latency by orders of magnitude as the data can fit in CPU caches"

FEBRUARY 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "it's a very clear well written coverage of all the things that are wrong with using finalize() and the alternatives available to you. If you are an advanced Java developer, you should read it"
• All the latest Java performance tips extracted in concise form
  • "A simple java.lang.ref.Cleaner example to clean up resources (for some obj) instead of implementing a finalize() method would be similar to Cleaner cleaner = Cleaner.create(); Cleaner.Cleanable cleanable = cleaner.register(obj, cleanupRunnableWhenObjGCed) and if you want to execute the clean up early, you can execute cleanable.clean() (which would execute the cleanupRunnableWhenObjGCed runnable immediately. The runnable is never executed more than once even including GCs"

JANUARY 2022 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "What's most interesting is how simple and straightforward the basic implementation of a Java simple agent is. If you want to try it, you can be up and running with your own agent implementation in minutes!"
• All the latest Java performance tips extracted in concise form
  • "The JVM has 3 entry points: agentmain, premain, main. -javaagent uses premain, which also runs on the main thread, and you can have multiple java agents. The agentmain is on it's own thread and if you attach to a JVM you enter this thread. You can have multiple agents but there is only one entry thread"

DECEMBER 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "The four golden signals are often co-dependent. A typical scenario is for saturation to be too high on some resource, causing errors to increase, throughput to start failing (decrease) and latency consequently to increase"
• All the latest Java performance tips extracted in concise form
  • "Virtual threads (aka fibres from project Loom) is more readable and efficient than reactive to multiplex, if you have sufficient memory for the number of virtual threads you want"

NOVEMBER 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "JVM GC logs analyzers from Oracle, HP, IBM, Netflix, Microsoft, Apple and many individual contributors tells us that the GC has been and still is an important target for tuning"
• All the latest Java performance tips extracted in concise form
  • "In the cloud, the footprint matters and is most likely the limiting factor in minimizing cost of a JVM application"

OCTOBER 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "GC log analysis tools are very useful - and now, one of the two commercial GC log analyzers, Censum, has open sourced it's core as GCToolkit."
• All the latest Java performance tips extracted in concise form
  • "Workloads have become memory bound - CPUs spend most of their time waiting for data to reach them, rather than processing that data. This means you often need to focus on optimizing getting data to the CPU to get speedups. Focus on "Instructions per cycle" - production workloads should be around 1.5 if efficient. Note that a stalled CPU is still reported as 100% utilized by the metric"

SEPTEMBER 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "There are a host of options and decisions to make when deciding on how available your application will be, each with consequences to your design, architecture and implementations. Make sure you decide on the availability you want to provide and ensure that you can attain it"
• All the latest Java performance tips extracted in concise form
  • "compare-and-swap is implemented equivalent to a volatile read and write operation on updates, but only a read operation (no write) if the compare fails. So it won't flush memory if the compare fails"

AUGUST 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "All the new performance features in the Java 17 release, the performance benefits from each feature, and how to use the feature"
• All the latest Java performance tips extracted in concise form
  • "To enable asynchronous logging of the JVM logs, use -Xlog:async and if you need to configure the logging buffer size, use eg -XX:AsyncLogBufferSize=1M (there is a dedicated thread for flushing the buffer)"

JULY 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "I explain how the three pillars of Observability (metrics, tracing, logs), fit into overall troubleshooting"
• All the latest Java performance tips extracted in concise form
  • "Shared state that is frequently written to by multiple threads should be avoided, as it fundamentally limits your scalability"

JUNE 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Scalability and latency are separate performance targets and you need to have separate targets defined so you know where to target your efforts. You can't just assume scalability can also give you your desired latencies."
• All the latest Java performance tips extracted in concise form
  • "System.nanotime can be 2 orders of magnitude slower depending on which timer source the OS uses"

MAY 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Health checks, metrics, telemetry, logs, tracing, profiling - how do they all fit together? I explain"
• All the latest Java performance tips extracted in concise form
  • "-Xgcpolicy:optthruput is a single generation parallel non-concurrent GC with optional compaction targeted at GC throughput; -Xgcpolicy:gencon is a parallel and concurrent 2-generation GC with concurrent collection in both generations; metronome is an incremental parallel GC with an upper bound on GC pause time; -Xgcpolicy:balanced is like optthruput but reduces the max pause time by doing smaller partial GCs where possible."

APRIL 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "I list the top 5 reasons for using APM"
• All the latest Java performance tips extracted in concise form
  • "Having balanced requests across instances is important, otherwise instances can be overloaded and you can have a cascaded failure of all instances."

MARCH 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "Being able to scale is challenging. Being able to autoscale needs you to understand what metrics determine when you need more (or fewer) resources allocated. Architecting to achieve a queueable application is fundamentally different from the ground up."
• All the latest Java performance tips extracted in concise form
  • "There are three options to handle as much traffic as possible, which can be used in combination: scale (adding sufficient additional capacity to handle the traffic increase); overprovision (already having sufficient additional capacity to handle any traffic increase); queue (temporarily holding requests somewhere and processing when resources are available)."

FEBRUARY 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "All the new performance features in the Java 16 release, the performance benefits from each feature, and how to use the feature - including the news that ZGC is fully concurrent - no pauses over a millisecond"
• All the latest Java performance tips extracted in concise form
  • "Unix domain sockets bypasses the TCP/IP stack with consequential improvements in latency and CPU usage"

JANUARY 2021 NEWSLETTER

• We list all the latest Java performance related news and articles
  • "From JDK 17 Shenandoah will be completely concurrent except for some constant time admin operations - all pauses be under 1 millisecond!"
• All the latest Java performance tips extracted in concise form
  • "Prioritized load shedding works by: Consistently prioritize requests across device types into NON_CRITICAL (does not affect customer experience), DEGRADED_EXPERIENCE (affects customer experience but doesn't prevent primary customer operations), CRITICAL, and score between 1 to 100 for each request given its individual characteristics; progressively throttle requests based on priority; validate by chaos testing (deliberate fault injection) for requests of specific priorities"

Previous newsletters

All our previous newsletters can be viewed from here

How to use this site

This site has four main information resources:

• The uncategorized tips page lists many other web pages with Java performance tuning related information. Each web page has its performance tuning tips extracted and listed immediately after the URL is listed. These tips are being categorized, and the tips page links to those categories currently available. If the performance area you are interested in is not yet categorized, send us an email from this page telling us the categorization you'd like to see. In any of the tips pages, use your browser's "find" or "search" option to identify particular tips you are interested in on the page, and follow up by reading the referenced web page if necessary or desired.
• The resources page lists Java performance tuning resources including books, tools, reports, other performance tuning sites of interest, and Java performance tuning discussion groups.
• The news pages are monthly newsletters listing any new Java performance tuning related information, together with Kirk Pepperdine's discussion group roundup and Javva The Hutt.
• The tool reports pages are detailed introductory reports on various Java performance related tools, for both free and commercial tools.

This site has been designed to download almost as fast as possible. (Some stylistic markup and required server-side processing has been used in preference to absolute speed contraints.) The web tree contains very few graphics and almost no complex typesetting markup except for very simple tables, and the main pages can be accessed directly from the menu.

This line is only to help search engines: Java performance tuning Java tuning Java optimization Java optimize Java optimizing Java fast Java faster Java speedup Java performance Java High-Performance