Benchmarking a slow machine

Overview

When performing benchmarks I usually reach for the fastest machine I can. The theory is that if speed matters you will use a fast machine.

Recently I have tried benchmarking the slowest machine I have access to. A Pentium IV dual core laptop with 4 GB of memory and a regular HDD. The assumption is that if this performs better than your need, it might not matter how fast your system is.

In the article, I am comparing this laptop to a fast machine. The point is, if the slow machine is more than enough you don't need to worry about hardware.

Latency test

The round trip average latencies tests were 10 times slower and there were over 1000x more delayed messages. Even so, a delay of an average of 4 micro-seconds is faster than many applications need. 
The average RTT latency was 3,442 ns. The 50/99 / 99.9/99.99%tile latencies 
    were 1,790/1,790 / 103,550/2,147,483,647. 
    There were 39,891 delays over 100 μs

Throughput test

The biggest difference I saw in the throughput test was that when the machine ran out memory, performance of the whole machine dropped dramatically. (Even the Task Manager stopped working) This is not too surprising as the performance becomes dependant on the speed of the HDD. The HDD in a budget laptop is slow even compared to that in a desktop. Compared with a fast SSD, it is not surprising that its much slower.

When testing the throughput performance for a dataset of 2.6 GB (the machine has 4 GB total) the performance is good, about four times slower. 

Took 10.454 seconds to write/read 60,000,000 entries, rate was 5.7 M entries/sec

Increase the dataset size to 5.2 GB and the performance drops to around 200 K entries/second. 

Took 974.180 seconds to write/read 200,000,000 entries, rate was 0.2 M entries/sec

Logging test

This logs short lines of text with the date, thread name a string and a double value. 
To log 1,000,000 messages took 0.631 seconds using Chronicle and 12.259 seconds using Logger
This is around three times slower than the fast machine.

Comments

  1. javin paul1 March 2012 at 14:26

    Good tip peter. We usually do performance benchmarking on production like system to gather correct metrics.

    Javin
    4 tips to improve performance of Java database application

    ReplyDelete
  2. Peter Lawrey1 March 2012 at 14:56

    A slow machine can be a good way to test your application under load (as its easier to over load)

    ReplyDelete

Post a Comment

Popular posts from this blog

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

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 ...
Read more

System wide unique nanosecond timestamps

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 ...
Read more

What does Chronicle Software do?

Image
Overview Chronicle Software is about simplifying fast data.  It is a suite of libraries to make it easier to write, monitor and tune data processing systems where performance and scalability are concerned. But its free, how do you make money, through support? We offer premium support . However, we often hear from users for the first time about a year after they have it in production.  Most users find they can support the software themselves.  It is only after a year or so, they have questions or concerns about where to take the product next.  In many cases, it is to a problem we have already solved and they just need to upgrade their software or use a new module we have added. As we don't know who is using most of our software, we can't advise them on how best to use our software and what updates or enhancements they would benefit from. We need users to contact us and ask questions.  We have a free forum , and we respond to 50% of questions in 2 hours...
Read more