Fleetbench

Fleetbench is a benchmarking suite for Google workloads. It's a portmanteau of "fleet" and "benchmark". It is meant for use by chip vendors, compiler researchers, and others interested in making performance optimizations beneficial to workloads similar to Google's. This repository contains the Fleetbench C++ code.

Details on Fleetbench can be found in our paper A Profiling-Based Benchmark Suite for Warehouse-Scale Computers.

NOTE: As this project is evolving, we recommend including the tag/release number when citing it to avoid any confusion.

Overview

Fleetbench is a benchmarking suite that consists of a curated set of microbenchmarks for hot functions across Google's fleet. The data set distributions it uses for executing the benchmarks are derived from data collected in production.

IMPORTANT: We released our latest v2.0.0, a major milestone that significantly enhances our benchmarking suite's capability to accurately characterize system performance under realistic, concurrent workloads. This release ensures the benchmark suite is runnable on emulation and real hardware, and introduces a powerful multiprocessing framework, alongside critical new benchmarks (gRPC and SIMD), and substantial improvements and bug fixes across the suite. Please check out the release note for more information!

This new version represents a substantial step forward in capturing system performance from diverse angles, enabling developers and performance engineers to gain granular insights into how important libraries behave in complex, multi-core environments.

For more information, see:

Workloads coverage for latest suite coverage.
Versioning on the details of Fleetbench's releases.
Running Benchmarks for how to run the benchmark.
Future Work section on all that we plan to add.

Benchmark fidelity

Benchmark fidelity is an important consideration in building this suite. There are 3 levels of fidelity that we consider:

The suite exercises the same functionality as production.
The suite's performance counters match production.
An optimization impact on the suite matches the impact on production.

Versioning

Fleetbench uses semantic versioning for its releases, where PATCH versions will be used for bug fixes, MINOR for updates to distributions and category coverage, and MAJOR for substantial changes to the benchmarking suite. All releases will be tagged, and the suite can be built and run at any version of the tag.

If you're starting out, authors recommend you always use the latest version at HEAD only.

Workloads coverage

As of Q2'25, Fleetbench provides coverage for several major hot libraries.

Benchmark	Description
Proto	Instruction-focused.
Swissmap	Data-focused.
Libc	Data-focused. Benchmarking `memcpy`, `memcmp/bcmp`, `memset`, and `memmove`.
TCMalloc	Data-focused.
Compression	Data-focused. Covers Snappy, ZSTD, Brotli, and Zlib.
Hashing	Data-focused. Supports algorithms CRC32 and absl::Hash.
STL-Cord	Instruction-focused.
RPC	Instruction-focused with a strong data-drive aspect and buit on gRPC framework
SIMD	ScaNN LUT16 based and measures performance of lookup-and-accumulate.

Benchmarks are classified by their core characteristics, such as being compute-bound, memory-bound, or sensitive to memory bandwidth vs. latency. For a detailed breakdown, see the benchmark characteristics documentation.

Running Benchmarks

Fleetbench supports running benchmarks in two modes: single-threaded and multi-cores. The following command is for a single-threaded run. For multi-core execution, please refer to the parallel run instructions.

Setup

Bazel is the official build system for Fleetbench.

We currently require Bazel version 8.0.0. Consider Bazelisk to automatically manage your Bazel version.

NOTE: Our setup uses LLVM 17.0.1.

As an example, to run the Swissmap benchmarks:

bazel run --config=opt fleetbench/swissmap:swissmap_benchmark

Important: Always run benchmarks with --config=opt to apply essential compiler optimizations.

Run commands

Replacing the $WORK_LOAD and $BUILD_TARGET with one of the entry in the table to build and run the benchmark. The reasons why we add each build flag are explained in the next few sections.

X86 {.new-tab}

bazel build --config=clang --config=opt \
--config=haswell fleetbench/WORK_LOAD:BUILD_TARGET
GLIBC_TUNABLES=glibc.pthread.rseq=0 bazel-bin/fleetbench/WORK_LOAD/BUILD_TARGET

Or combining build and run together:

GLIBC_TUNABLES=glibc.pthread.rseq=0 bazel run --config=clang --config=opt \
--config=haswell fleetbench/WORK_LOAD:BUILD_TARGET

Arm {.new-tab}

bazel build --config=clang --config=opt \
--config=arm fleetbench/WORK_LOAD:BUILD_TARGET
GLIBC_TUNABLES=glibc.pthread.rseq=0 bazel-bin/fleetbench/WORK_LOAD/BUILD_TARGET

Or combining build and run together:

GLIBC_TUNABLES=glibc.pthread.rseq=0 bazel run --config=clang --config=opt \
--config=arm fleetbench/WORK_LOAD:BUILD_TARGET

Benchmark	WORKLOAD	BUILD_TARGET	Binary run flags
Proto	proto	proto_benchmark
Swissmap	swissmap	swissmap_benchmark
Libc memory	libc	mem_benchmark	`--benchmark_counters_tabular=true`
TCMalloc	tcmalloc	empirical_driver	Check `--benchmark_filter` below.
Compression	compression	compression_benchmark	`--benchmark_counters_tabular=true`
Hashing	hashing	hashing_benchmark	`--benchmark_counters_tabular=true`
STL-Cord	stl	cord_benchmark
RPC	rpc	rpc_benchmark
SIMD	simd	simd_benchmark

NOTE: By default, each benchmark only runs a minimal set of tests that we have selected as the most representative. To see the default lists, you can use the --benchmark_list_tests flag when running the target. You can add --benchmark_filter=all to see the exhaustive list.

You can also specify a regex in --benchmark_filter flag to specify a subset of benchmarks to run (more info). The TCMalloc Empirical Driver benchmark can take ~1hr to run all benchmarks, so running a subset may be advised.

Example to run for only sets of 16 and 64 elements of swissmap:

bazel run --config=opt fleetbench/swissmap:swissmap_benchmark -- \
--benchmark_filter=".*set_size:(16|64).*"

To extend the runtime of a benchmark, e.g. to collect more profile samples, use --benchmark_min_time.

bazel run --config=opt fleetbench/proto:proto_benchmark -- --benchmark_min_time=30s

Some benchmarks also provide counter reports after completion. Adding --benchmark_counters_tabular=true (doc) can help print counters as table columns for improved layout.

Ensuring TCMalloc per-CPU Mode

TCMalloc is the underlying memory allocator in this benchmark suite. By default it operates in per-CPU mode.

Note: the Restartable Sequences (RSEQ) kernel feature is required for per-CPU mode. RSEQ has the limitation that a given thread can only register a single rseq structure with the kernel. Recent versions of glibc do this on initialization, preventing TCMalloc from using it.

Set the environment variable: GLIBC_TUNABLES=glibc.pthread.rseq=0 to prevent glibc from doing this registration. This will allow TCMalloc to operate in per-CPU mode.

Clang Toolchain

For more consistency with Google's build configuration, we suggest using the Clang / LLVM tools. These instructions have been tested with LLVM 14.

These can be installed with the system's package manager, e.g. on Debian:

sudo apt-get install clang llvm lld

Otherwise, see https://releases.llvm.org to obtain these if not present on your system or to find the newest version.

Once installed, specify --config=clang on the bazel command line to use the clang compiler. We assume clang and lld are in the PATH.

Note: to make this setting the default, add build --config=clang to your .bazelrc.

Architecture-Specific Flags

If running on an x86 Haswell or above machine, we suggest adding --config=haswell for consistency with our compiler flags.

Use --config=westmere for Westmere-era processors, and --config=arm for ARM ones.

Reducing run-to-run variance

It is expected that there will be some variance in the reported CPU times across benchmark executions. The benchmark itself runs the same code, so the causes of the variance are mainly in the environment. The following is a non-exhaustive list of techniques that help with reducing run-to-run latency variance:

Ensure no other workloads are running on the machine at the same time. Note that this makes the environment less representative of production, where multi-tenant workloads are common.
Run the benchmark for longer, controlled with --benchmark_min_time.
Run multiple repetitions of the benchmarks in one go, controlled with --benchmark_repetitions.
Recommended by the benchmarking framework here:
- Disable frequency scaling
- Bind the process to a core by setting its affinity
- Disable processor boosting
- Disable Hyperthreading/SMT (should not affect single-threaded benchmarks)
- NOTE: We do not recommend reducing the working set of the benchmark to fit into L1 cache, contrary to the recommendations in the link, as it would significantly reduce this benchmarking suite's representativeness.
Disable memory randomization (ASLR)

Future Work

Potential areas of future work include:

Increasing the set of benchmarks included in the suite to capture more of the range of code executed by google workloads.
Generate benchmarking score.
Update data distributions based on new fleet measurements.
Rewrite individual components with macrobenchmarks.
Extend the benchmarking suite to allow for drop-in replacement of equivalent implementations for each category of workloads.

FAQs

Q: How do I compare results of the two different runs of a benchmark, e.g. contender vs baseline?

A: Fleetbench is using the benchmark framework. Please reference its documentation for comparing results across benchmark runs: link.
Q: How do I build the benchmark with FDO?

A: Note that Clang and the LLVM tools are required for FDO builds.

Take fleetbench/swissmap/swissmap_benchmark as an example.

# Instrument.
bazel build --config=clang --config=opt --fdo_instrument=.fdo fleetbench/swissmap:swissmap_benchmark
# Run to generate instrumentation.
bazel-bin/fleetbench/swissmap/swissmap_benchmark --benchmark_filter=all
# There should be a file with a .profraw extension in $PWD/.fdo/.
# Build an optimized binary.
bazel build --config=clang --config=opt --fdo_optimize=.fdo/<filename>.profraw fleetbench/swissmap:swissmap_benchmark
# Run the FDO-optimized binary.
bazel-bin/fleetbench/swissmap/swissmap_benchmark --benchmark_filter=all

Q: How do I build the benchmark with ThinLTO?

A: Note that Clang and the LLVM tools are required for ThinLTO builds. In particular, the lld linker must be in the PATH. Specify --features=thin_lto on the bazel command line. E.g.

bazel run --config=clang --config=opt --features=thin_lto fleetbench/proto:proto_benchmark

Q: Does Fleetbench run on _ OS?

A: The supported platforms are same as TCMalloc's, see link for more details.
Q: Can I run Fleetbench without TCMalloc?

A: Yes. Specify --custom_malloc="@bazel_tools//tools/cpp:malloc" on the bazel command line to override with the system allocator.
Q: Can I run with Address Sanitizer?

A: Yes. Note that you need to override TCMalloc as well for ASAN to work.

Example:

    bazel build --custom_malloc="@bazel_tools//tools/cpp:malloc" -c opt fleetbench/proto:proto_benchmark --copt=-fsanitize=address --linkopt=-fsanitize=address

Q: Are the benchmarks fixed in nature?

A: No. It is our expectation that the code under benchmark, the hardware, the compiler, and compiler flags used may all change in concert as to identify optimization opportunities.
Q: My question isn't addressed here. How do I contact the development team?

A: Please see previous GH issues and file a new one, if your question isn't addressed there.

License

Fleetbench is licensed under the terms of the Apache license. See LICENSE for more information.

Disclaimer: This is not an officially supported Google product.