Profiling Particular person Queries in a Concurrent System

A superb CPU profiler is value its weight in gold. Measuring efficiency in-situ often means utilizing a sampling profile. They supply quite a lot of info whereas having very low overhead. In a concurrent system, nonetheless, it’s exhausting to make use of the ensuing information to extract high-level insights. Samples don’t embody context like question IDs and application-level statistics; they present you what code was run, however not why.

This weblog introduces trampoline histories, a method Rockset has developed to effectively connect application-level info (question IDs) to the samples of a CPU profile. This lets us use profiles to grasp the efficiency of particular person queries, even when a number of queries are executing concurrently throughout the identical set of employee threads.

Primer on Rockset

Rockset is a cloud-native search and analytics database. SQL queries from a buyer are executed in a distributed vogue throughout a set of servers within the cloud. We use inverted indexes, approximate vector indexes, and columnar layouts to effectively execute queries, whereas additionally processing streaming updates. Nearly all of Rockset’s performance-critical code is C++.

Most Rockset prospects have their very own devoted compute assets known as digital cases. Inside that devoted set of compute assets, nonetheless, a number of queries can execute on the identical time. Queries are executed in a distributed vogue throughout the entire nodes, so which means that a number of queries are energetic on the identical time in the identical course of. This concurrent question execution poses a problem when making an attempt to measure efficiency.

Concurrent question processing improves utilization by permitting computation, I/O, and communication to be overlapped. This overlapping is very vital for top QPS workloads and quick queries, which have extra coordination relative to their elementary work. Concurrent execution can also be vital for decreasing head-of-line blocking and latency outliers; it prevents an occasional heavy question from blocking completion of the queries that comply with it.

We handle concurrency by breaking work into micro-tasks which are run by a hard and fast set of thread swimming pools. This considerably reduces the necessity for locks, as a result of we are able to handle synchronization by way of activity dependencies, and it additionally minimizes context switching overheads. Sadly, this micro-task structure makes it troublesome to profile particular person queries. Callchain samples (stack backtraces) might need come from any energetic question, so the ensuing profile exhibits solely the sum of the CPU work.

Profiles that mix the entire energetic queries are higher than nothing, however quite a lot of handbook experience is required to interpret the noisy outcomes. Trampoline histories allow us to assign many of the CPU work in our execution engine to particular person question IDs, each for steady profiles and on-demand profiles. This can be a very highly effective software when tuning queries or debugging anomalies.

DynamicLabel

The API we’ve constructed for including application-level metadata to the CPU samples is named DynamicLabel. Its public interface could be very easy:

class DynamicLabel {
  public:
    DynamicLabel(std::string key, std::string worth);
    ~DynamicLabel();

    template <typename Func>
    std::invoke_result_t<Func> apply(Func&& func) const;
};

DynamicLabel::apply invokes func. Profile samples taken throughout that invocation can have the label hooked up.

Every question wants just one DynamicLabel. Each time a micro-task from the question is run it’s invoked by way of DynamicLabel::apply.

One of the vital vital properties of sampling profilers is that their overhead is proportional to their sampling fee; that is what lets their overhead be made arbitrarily small. In distinction, DynamicLabel::apply should do some work for each activity whatever the sampling fee. In some circumstances our micro-tasks might be fairly micro, so it is vital that apply has very low overhead.

apply‘s efficiency is the first design constraint. DynamicLabel‘s different operations (development, destruction, and label lookup throughout sampling) occur orders of magnitude much less often.

Let’s work by way of some methods we would attempt to implement the DynamicLabel performance. We’ll consider and refine them with the objective of constructing apply as quick as potential. If you wish to skip the journey and soar straight to the vacation spot, go to the “Trampoline Histories” part.

Implementation Concepts

Concept #1: Resolve dynamic labels at pattern assortment time

The obvious strategy to affiliate utility metadata with a pattern is to place it there from the start. The profiler would search for dynamic labels on the identical time that it’s capturing the stack backtrace, bundling a replica of them with the callchain.

Rockset’s profiling makes use of Linux’s perf_event, the subsystem that powers the perf command line software. perf_event has many benefits over signal-based profilers (comparable to gperftools). It has decrease bias, decrease skew, decrease overhead, entry to {hardware} efficiency counters, visibility into each userspace and kernel callchains, and the flexibility to measure interference from different processes. These benefits come from its structure, during which system-wide profile samples are taken by the kernel and asynchronously handed to userspace by way of a lock-free ring buffer.

Though perf_event has quite a lot of benefits, we are able to’t use it for thought #1 as a result of it could possibly’t learn arbitrary userspace information at sampling time. eBPF profilers have the same limitation.

Concept #2: File a perf pattern when the metadata modifications

If it’s not potential to tug dynamic labels from userspace to the kernel at sampling time, then what about push? We might add an occasion to the profile each time that the thread→label mapping modifications, then post-process the profiles to match up the labels.

A method to do that could be to make use of perf uprobes. Userspace probes can document operate invocations, together with operate arguments. Sadly, uprobes are too gradual to make use of on this vogue for us. Thread pool overhead for us is about 110 nanoseconds per activity. Even a single crossing from the userspace into the kernel (uprobe or syscall) would multiply this overhead.

Avoiding syscalls throughout DynamicLabel::apply additionally prevents an eBPF resolution, the place we replace an eBPF map in apply after which modify an eBPF profiler like BCC to fetch the labels when sampling.

Concept #3: Merge profiles with a userspace label historical past

If it is too costly to document modifications to the thread→label mapping within the kernel, what if we do it within the userspace? We might document a historical past of calls to DynamicLabel::apply, then be a part of it to the profile samples throughout post-processing. perf_event samples can embody timestamps and Linux’s CLOCK_MONOTONIC clock has sufficient precision to seem strictly monotonic (at the very least on the x86_64 or arm64 cases we would use), so the be a part of could be actual. A name to clock_gettime utilizing the VDSO mechanism is lots sooner than a kernel transition, so the overhead could be a lot decrease than that for thought #2.

The problem with this method is the info footprint. DynamicLabel histories could be a number of orders of magnitude bigger than the profiles themselves, even after making use of some easy compression. Profiling is enabled repeatedly on all of our servers at a low sampling fee, so making an attempt to persist a historical past of each micro-task invocation would rapidly overload our monitoring infrastructure.

Concept #4: In-memory historical past merging

The sooner we be a part of samples and label histories, the much less historical past we have to retailer. If we might be a part of the samples and the historical past in near-realtime (maybe each second) then we wouldn’t want to write down the histories to disk in any respect.

The most typical method to make use of Linux’s perf_event subsystem is by way of the perf command line software, however the entire deep kernel magic is on the market to any course of by way of the perf_event_open syscall. There are quite a lot of configuration choices (perf_event_open(2) is the longest manpage of any system name), however when you get it arrange you possibly can learn profile samples from a lock-free ring buffer as quickly as they’re gathered by the kernel.

To keep away from competition, we might keep the historical past as a set of thread-local queues that document the timestamp of each DynamicLabel::apply entry and exit. For every pattern we’d search the corresponding historical past utilizing the pattern’s timestamp.

This method has possible efficiency, however can we do higher?

Concept #5: Use the callchains to optimize the historical past of calls to `apply`

We will use the truth that apply exhibits up within the recorded callchains to cut back the historical past measurement. If we block inlining in order that we are able to discover DynamicLabel::apply within the name stacks, then we are able to use the backtrace to detect exit. Which means apply solely wants to write down the entry data, which document the time that an affiliation was created. Halving the variety of data halves the CPU and information footprint (of the a part of the work that’s not sampled).

This technique is the most effective one but, however we are able to do even higher! The historical past entry data a variety of time for which apply was sure to a specific label, so we solely have to make a document when the binding modifications, moderately than per-invocation. This optimization might be very efficient if now we have a number of variations of apply to search for within the name stack. This leads us to trampoline histories, the design that now we have carried out and deployed.

Trampoline Histories

If the stack has sufficient info to seek out the precise DynamicLabel , then the one factor that apply must do is go away a body on the stack. Since there are a number of energetic labels, we’ll want a number of addresses.

A operate that instantly invokes one other operate is a trampoline. In C++ it would seem like this:

__attribute__((__noinline__))
void trampoline(std::move_only_function<void()> func) {
    func();
    asm unstable (""); // stop tailcall optimization
}

Word that we have to stop compiler optimizations that may trigger the operate to not be current within the stack, particularly inlining and tailcall elimination.

The trampoline compiles to solely 5 directions, 2 to arrange the body pointer, 1 to invoke func(), and a pair of to wash up and return. Together with padding that is 32 bytes of code.

C++ templates allow us to simply generate a complete household of trampolines, every of which has a novel tackle.

utilizing Trampoline = __attribute__((__noinline__)) void (*)(
        std::move_only_function<void()>);

constexpr size_t kNumTrampolines = ...;

template <size_t N>
__attribute__((__noinline__))
void trampoline(std::move_only_function<void()> func) {
    func();
    asm unstable (""); // stop tailcall optimization
}

template <size_t... Is>
constexpr std::array<Trampoline, sizeof...(Is)> makeTrampolines(
        std::index_sequence<Is...>) {
    return {&trampoline<Is>...};
}

Trampoline getTrampoline(unsigned idx) {
    static constexpr auto kTrampolines =
            makeTrampolines(std::make_index_sequence<kNumTrampolines>{});
    return kTrampolines.at(idx);
}

We’ve now bought the entire low-level items we have to implement DynamicLabel:

• DynamicLabel development → discover a trampoline that’s not at the moment in use, append the label and present timestamp to that trampoline’s historical past
• DynamicLabel::apply → invoke the code utilizing the trampoline
• DynamicLabel destruction → return the trampoline to a pool of unused trampolines
• Stack body symbolization → if the trampoline’s tackle is present in a callchain, lookup the label within the trampoline’s historical past

Efficiency Affect

Our objective is to make DynamicLabel::apply quick, in order that we are able to use it to wrap even small items of labor. We measured it by extending our present dynamic thread pool microbenchmark, including a layer of indirection by way of apply.

{
    DynamicThreadPool executor({.maxThreads = 1});
    for (size_t i = 0; i < kNumTasks; ++i) {
        executor.add([&]() {
            label.apply([&] { ++rely; }); });
    }
    // ~DynamicThreadPool waits for all duties
}
EXPECT_EQ(kNumTasks, rely);

Maybe surprisingly, this benchmark exhibits zero efficiency affect from the additional degree of indirection, when measured utilizing both wall clock time or cycle counts. How can this be?

It seems we’re benefiting from a few years of analysis into department prediction for oblique jumps. The within of our trampoline seems like a digital technique name to the CPU. That is extraordinarily widespread, so processor distributors have put quite a lot of effort into optimizing it.

If we use perf to measure the variety of directions within the benchmark we observe that including label.apply causes about three dozen further directions to be executed per loop. This could gradual issues down if the CPU was front-end sure or if the vacation spot was unpredictable, however on this case we’re reminiscence sure. There are many execution assets for the additional directions, so that they don’t truly enhance this system’s latency. Rockset is mostly reminiscence sure when executing queries; the zero-latency consequence holds in our manufacturing surroundings as effectively.

A Few Implementation Particulars

There are some things we have executed to enhance the ergonomics of our profile ecosystem:

• The perf.information format emitted by perf is optimized for CPU-efficient writing, not for simplicity or ease of use. Although Rockset’s perf_event_open-based profiler pulls information from perf_event_open, now we have chosen to emit the identical protobuf-based pprof format utilized by gperftools. Importantly, the pprof format helps arbitrary labels on samples and the pprof visualizer already has the flexibility to filter on these tags, so it was simple so as to add and use the knowledge from DynamicLabel.
• We subtract one from most callchain addresses earlier than symbolizing, as a result of the return tackle is definitely the primary instruction that might be run after returning. That is particularly vital when utilizing inline frames, since neighboring directions are sometimes not from the identical supply operate.
• We rewrite trampoline<i> to trampoline<0> in order that now we have the choice of ignoring the tags and rendering an everyday flame graph.
• When simplifying demangled constructor names, we use one thing like Foo::copy_construct and Foo::move_construct moderately than simplifying each to Foo::Foo. Differentiating constructor sorts makes it a lot simpler to seek for pointless copies. (Should you implement this ensure you can deal with demangled names with unbalanced < and >, comparable to std::enable_if<sizeof(Foo) > 4, void>::sort.)
• We compile with -fno-omit-frame-pointer and use body tips that could construct our callchains, however some vital glibc capabilities like memcpy are written in meeting and don’t contact the stack in any respect. For these capabilities, the backtrace captured by perf_event_open‘s PERF_SAMPLE_CALLCHAIN mode omits the operate that calls the meeting operate. We discover it through the use of PERF_SAMPLE_STACK_USER to document the highest 8 bytes of the stack, splicing it into the callchain when the leaf is in a kind of capabilities. That is a lot much less overhead than making an attempt to seize your entire backtrace with PERF_SAMPLE_STACK_USER.

Conclusion

Dynamic labels let Rockset tag CPU profile samples with the question whose work was energetic at that second. This potential lets us use profiles to get insights about particular person queries, despite the fact that Rockset makes use of concurrent question execution to enhance CPU utilization.

Trampoline histories are a method of encoding the energetic work within the callchain, the place the present profiling infrastructure can simply seize it. By making the DynamicLabel ↔ trampoline binding comparatively long-lived (milliseconds, moderately than microseconds), the overhead of including the labels is saved extraordinarily low. The method applies to any system that desires to reinforce sampled callchains with utility state.

Rockset is hiring engineers in its Boston, San Mateo, London and Madrid places of work. Apply to open engineering positions at the moment.