Unexpected Full GCs Triggered by RMI in Latency-Sensitive Applications

By

We observed an unexpected increase in Full Garbage Collections (Full GCs) while optimising a latency-sensitive application with minimal object creation. Despite reducing the frequency of minor GCs to enhance performance, the system began to exhibit hourly periodic pauses due to Full GCs, which was counterintuitive.

Investigating the Source of Full GCs

Upon closer examination, we discovered that the Java Remote Method Invocation (RMI) system was initiating Full GCs every hour. Specifically, the RMI Distributed Garbage Collector (DGC) checks if a GC has occurred in the last hour and, if not, forces a Full GC. This behaviour occurs even if the application does not actively use RMI, leading to unnecessary performance overhead.

Understanding RMI's Impact on Garbage Collection

The RMI DGC collects periodic garbage to clean up unused remote objects. By default, it is configured to trigger a Full GC if none has occurred within a specified interval (defaulting to one hour). This mechanism ensures that stale references are collected in distributed applications. However, for applications that do not use RMI or where latency is critical, this can introduce unwelcome Full GCs.

Configuring RMI DGC Intervals

To mitigate this issue, we can adjust the intervals at which the RMI DGC checks for garbage collection. This is done by setting the following JVM options:

-Dsun.rmi.dgc.client.gcInterval=86400000
-Dsun.rmi.dgc.server.gcInterval=86400000

The values are in milliseconds, so setting them to 86400000 extends the interval to 24 hours. It's important to note that the client and server intervals must be increased; adjusting only one has no effect due to how RMI DGC schedules its checks.

Applying the Solution

By updating these JVM options, we observed a significant reduction in unnecessary Full GCs. Our application's latency improved as the unexpected pauses caused by the RMI DGC were eliminated.

Performance Considerations

While extending the RMI DGC intervals mitigates the issue for non-RMI applications, be cautious if your application uses RMI. Increasing the intervals might delay the cleanup of unused remote objects, potentially leading to memory leaks or other resource management issues.

Conclusion

Unanticipated behaviours like RMI-induced Full GCs can impact application performance, especially in latency-sensitive systems. You must be aware of default JVM settings that may not align with your application's needs. Adjusting the sun.rmi.dgc.client.gcInterval and sun.rmi.dgc.server.gcInterval options can help optimise garbage collection behaviour.

Have You Experienced Similar Issues?

Have you encountered unexpected Full GCs in your applications? How did you address them? Share your experiences and insights below.

About the Author

Peter Lawrey is the CEO of Chronicle Software

Comments

Post a Comment

Popular posts from this blog

Java is Very Fast, If You Don’t Create Many Objects

By
Image
  You still have to watch how many objects you create. This article looks at a benchmark passing events over TCP/IP at 4 billion events per minute using the net.openhft.chronicle.wire.channel package in Chronicle Wire and why we still avoid object allocations..  One of the key optimisations is creating almost no garbage. Allocation is a very cheap operation and collection of very short-lived objects is also very cheap. Does this really make a difference? What difference does one small object per event (44 bytes) make to the performance in a throughput test where GC pauses are amortised? While allocation is as efficient as possible, it doesn’t avoid the memory pressure on the L1/L2 caches of your CPUs and when many cores are busy, they are contending for memory in the shared L3 cache.  Results Benchmark on a Ryzen 5950X with Ubuntu 22.10. JVM Vendor, Version No objects Throughput, Average Latency* One object per event Throughput, Average Latency* Azul Zulu 1.8.0_322 60.6 M event/s, 528
Read more

System wide unique nanosecond timestamps

By
A Unique Identifier can be very useful for tracing. Those ids are even more useful when they contain a high-resolution timestamp.  Not only do they record the time of an event, but if unique can help trace events as they pass through the system. Such unique timestamps however can be expensive depending on how they are implemented.   This post explores a lightweight means of producing a unique, monotonically increasing system-wide nano-second resolution timestamp available in our open-source library. Uses for Unique Identifiers Unique identifiers can be useful to associate with a piece of information so that information can be referred to later unambiguously. This could be an event, a request, an order id, or a customer id. They can naturally be used as a primary key in a database or key/value store to retrieve that information later. One of the challenges of generating these identifiers is avoiding creating duplicates while not having an increasing cost.  You could record every identi
Read more

Comparing Approaches to Durability in Low Latency Messaging Queues

By
Image
  A significant feature of Chronicle Queue Enterprise is support for TCP replication across multiple servers to ensure the high availability of application infrastructure. I generally believe that replicating data to a secondary system is faster than syncing to disk, assuming the round trip network delay wasn’t high due to quality networks and co-located redundant servers. This is the first time I have benchmarked it with a realistic example. Little’s Law and Why Latency Matters In many cases, the assumption is that the latency won't be a problem as long as throughput is high enough. However, latency is often a key factor in why the throughput isn’t high enough. Little’s law states, “ the long-term average number  L  of customers in a  stationary  system is equal to the long-term average effective arrival rate  λ  multiplied by the average time  W  that a customer spends in the system”. In computer terminology, the level of concurrency or parallelism a system has to support must be
Read more