Brendan Gregg

JUNE 30, 2018

These have inspired me to summarize another performance activity: evaluating benchmark accuracy. Accurate benchmarking rewards engineering investment that actually improves performance, but, unfortunately, inaccurate benchmarking is more common. If the benchmark reported 20k ops/sec, you should ask: why not 40k ops/sec?

Benchmarking 47

Benchmarking Latency Cache Network 47

Scalegrid

MARCH 28, 2024

This article will explore how they handle data storage and scalability, perform in different scenarios, and, most importantly, how these factors influence your choice. It uses a hash table to manage these pairs, divided into fixed-size buckets with linked lists for key-value storage. Redis Database Management with ScaleGrid ScaleGrid.io

Cache 130

Cache Storage Architecture Scalability 130

Percona

DECEMBER 26, 2023

Querying the data While it is reasonable to create panels showing real-time load in order to explore better the types of queries that can be run against pg_stat_monitor, it is more practical to copy and query the data into tables after the benchmarking has completed its run. A script executing a benchmarking run: #!/bin/bash

Benchmarking 126

Benchmarking Metrics C++ Database 126

Brendan Gregg

JUNE 29, 2018

These have inspired me to summarize another performance activity: evaluating benchmark accuracy. Accurate benchmarking rewards engineering investment that actually improves performance, but, unfortunately, inaccurate benchmarking is more common. If the benchmark reported 20k ops/sec, you should ask: why not 40k ops/sec?

Benchmarking 40

Benchmarking Latency Cache Network 40

The Netflix TechBlog

OCTOBER 19, 2020

If we had an ID for each streaming session then distributed tracing could easily reconstruct session failure by providing service topology, retry and error tags, and latency measurements for all service calls. Our distributed tracing infrastructure is grouped into three sections: tracer library instrumentation, stream processing, and storage.

Infrastructure 236

Infrastructure Transportation Storage Open Source 236

The Netflix TechBlog

SEPTEMBER 2, 2020

This difference has substantial technological implications, from the classification of what’s interesting to transport to cost-effective storage (keep an eye out for later Netflix Tech Blog posts addressing these topics). As you can imagine, this comes with very real storage costs. Is this an anomaly or are we dealing with a pattern?

Latency 298

Latency Transportation Engineering Traffic 298

The Netflix TechBlog

NOVEMBER 22, 2019

4:45pm-5:45pm NFX 209 File system as a service at Netflix Kishore Kasi , Senior Software Engineer Abstract : As Netflix grows in original content creation, its need for storage is also increasing at a rapid pace. In order to maintain performance, benchmarking is a vital part of our system’s lifecycle.

AWS 100

AWS Entertainment Open Source Benchmarking 100

The Netflix TechBlog

NOVEMBER 22, 2019

4:45pm-5:45pm NFX 209 File system as a service at Netflix Kishore Kasi , Senior Software Engineer Abstract : As Netflix grows in original content creation, its need for storage is also increasing at a rapid pace. In order to maintain performance, benchmarking is a vital part of our system’s lifecycle.

AWS 100

AWS Entertainment Open Source Benchmarking 100

Scalegrid

JANUARY 12, 2024

Storage is a critical aspect to consider when working with cloud workloads. High availability storage options within the context of cloud computing involve highly adaptable storage solutions specifically designed for storing vast amounts of data while providing easy access to it. What is an example of a workload?

Cloud 130

Cloud Virtualization Storage Efficiency 130

Particular Software

SEPTEMBER 20, 2021

Some will claim that any type of RPC communication ends up being faster (meaning it has lower latency) than any equivalent invocation using asynchronous messaging. If you did such a benchmark, here’s an incomplete picture you might end up with: Graph of microbenchmark showing RPC is faster than messaging. Messaging doesn’t do that.

Benchmarking 97

Benchmarking Latency Servers Systems 97

Percona

MARCH 23, 2023

Storage The type of storage and disk used for database servers can have a significant impact on performance and reliability. Benchmark before you decide. Cloud Different cloud providers offer a range of instance types and sizes, each with varying amounts of CPU, memory, and storage. Transparent huge pages (THP) disabled.

Performance 129

Performance Hardware Tuning Storage 129

The Morning Paper

NOVEMBER 8, 2020

The initial version of Delos went into production after eight months using a ZooKeeper-backed Loglet implementation, and then four months later it was swapped out for a new custom-built NativeLoglet that gave a 10x improvement in end-to-end latency. replacing Paxos with Raft), or they could be shims over external storage systems.

Virtualization 126

Virtualization Latency Storage Systems 126

Percona

JUNE 22, 2023

Indexing efficiency Monitoring indexing efficiency in MySQL involves analyzing query performance, using EXPLAIN statements, utilizing performance monitoring tools, reviewing error logs, performing regular index maintenance, and benchmarking/testing. This KPI is also directly related to Query Performance and helps improve it.

Performance 109

Performance Monitoring Traffic Database 109

Particular Software

OCTOBER 26, 2020

Backed by Cosmos DB, a fully managed, globally distributed, elastically scaled, pay-as-you-go service, your NServiceBus-based systems can benefit from guaranteed single-digit-millisecond latency with 99.999% availability. How does this compare with Azure Storage Persistence?

Azure 52

Azure Storage Benchmarking Latency 52

n0derunner

SEPTEMBER 17, 2018

Creating a HCI benchmark to simulate multi-tennent workloads. As with traditional storage, applications are writing to a shared storage environment which is necessary to support VM movement. As with traditional storage, applications are writing to a shared storage environment which is necessary to support VM movement.

Performance Testing 40

Performance Testing Benchmarking Testing Performance 40

The Morning Paper

AUGUST 29, 2019

use the TPC-H benchmark to assess Redshift, Redshift Spectrum, Athena, Presto, Hive, and Vertica to find out what works best and the trade-offs involved. For cost calculations, the costs are a combination of compute costs, storage costs, data scan costs, and software license costs. Key findings. System initialisation time.

Architecture 76

Architecture Cloud Storage Serverless 76

Percona

APRIL 7, 2023

To illustrate this, I ran the Sysbench-TPCC synthetic benchmark against two different GCP instances running a freshly installed Percona Server for MySQL version 8.0.31 In MySQL, considering the standard storage engine, InnoDB , the data cache is called Buffer Pool. MySQL (B) 2517529 2610323 389048 5516900 194140 11523.48

Tuning 139

Tuning Cache Servers Benchmarking 139

HammerDB

NOVEMBER 5, 2018

This post at an entry-level discusses the options you have to improve log throughput in your benchmark environment. . Additionally for the log disk component it is latency for an individual write that is crucial rather than the total I/O bandwidth. innodb_flush_log_at_trx_commit=0. PostgreSQL.

Benchmarking 40

Benchmarking Database C++ Virtualization 40

Percona

JANUARY 24, 2023

This will be clearly visible in PostgreSQL performance benchmarks as a “ Sawtooth wave ” pattern observed by Vadim in his tests: As we can see, the throughput suddenly drops after every checkpoint due to heavy WAL writing and gradually picks up until the next checkpoint. But this comes with a considerable performance implication.

Database 121

Database Benchmarking Open Source Latency 121

SQL Performance

APRIL 8, 2019

Back on December 5, 2017, Microsoft announced that they were using AMD EPYC 7551 processors in their storage-optimized Lv2-Series virtual machines. I wrote about using CPU-Z to benchmark the Intel Xeon E5-2673 v3 processor in an Azure VM in this article. Figure 1: CPU-Z Benchmark Results for LS16v2. Memory (GiB).

Azure 42

Azure Virtualization Storage Benchmarking 42

Percona

JULY 1, 2023

It efficiently manages read and write operations, optimizes data access, and minimizes contention, resulting in high throughput and low latency to ensure that applications perform at their best. While RDS MySQL supports multiple storage engines with varying capabilities, not all of them are optimized for crash recovery and data durability.

AWS 52

AWS Database Serverless Storage 52

Abhishek Tiwari

DECEMBER 27, 2017

faster access to external storage and data locality (I/O, bandwidth). A recent performance benchmark completed by Intel and BlueData using the BigBench benchmarking kit has shown that the performance ratios for container-based Hadoop workloads on BlueData EPIC are equal to and in some cases, better than bare-metal Hadoop [7].

Big Data 63

Big Data Storage Benchmarking Hardware 63

HammerDB

DECEMBER 5, 2020

The HammerDB TPROC-C workload by design intended as CPU and memory intensive workload derived from TPC-C – so that we get to benchmark at maximum CPU performance at a much smaller database footprint. For TPC-C this meant enough available spindles to reduce I/O latency and for TPC-H enough bandwidth for data throughput.

C++ 40

C++ Benchmarking Virtualization Open Source 40

The Netflix TechBlog

NOVEMBER 22, 2019

4:45pm-5:45pm NFX 209 File system as a service at Netflix Kishore Kasi , Senior Software Engineer Abstract : As Netflix grows in original content creation, its need for storage is also increasing at a rapid pace. In order to maintain performance, benchmarking is a vital part of our system’s lifecycle.

AWS 38

AWS Entertainment Open Source Benchmarking 38

Smashing Magazine

SEPTEMBER 28, 2022

On your first try, you can use it as a benchmark for optimizations later. Caching partially stores your data and is not used as permanent storage. Using the cache as permanent storage is an anti-pattern. At the bottom of the page, you can check how many kB of resources your client consumes. Caching Schemes. Large preview ).

Cache 128

Cache Performance Servers Architecture 128

Percona

SEPTEMBER 1, 2023

This reduction in latency ensures that applications and websites provide a more rapid and responsive user experience. By analyzing disk I/O metrics, you can optimize queries to reduce disk reads or upgrade to faster storage solutions. Avoid over-indexing, which can bloat storage and slow writes.

Tuning 52

Tuning Database Performance Hardware 52

Alex Russell

MARCH 6, 2021

A then-representative $200USD device had 4-8 slow (in-order, low-cache) cores, ~2GiB of RAM, and relatively slow MLC NAND flash storage. Using a global ASP as a benchmark can further mislead thanks to the distorting effect of ultra-high-end prices rising while shipment volumes stagnate. The Moto G4 , for example.

Performance 127

Performance Network Cache Metrics 127

SQL Server According to Bob

SEPTEMBER 26, 2016

When we released Always On Availability Groups in SQL Server 2012 as a new and powerful way to achieve high availability, hardware environments included NUMA machines with low-end multi-core processors and SATA and SAN drives for storage (some SSDs). This chart shows our scaled results using a OLTP workload derived from TPC benchmarks.

Availability 40

Availability Servers Hardware Benchmarking 40

Alex Russell

OCTOBER 22, 2017

Budgets are scaled to a benchmark network & device. Deciding what benchmark to use for a performance budget is crucial. It simulates a link with a 400ms RTT and 400-600Kbps of throughput (plus latency variability and simulated packet loss). Performance budgets are set early in the life of the project. Global Ground-Truth.

Performance 74

Performance Benchmarking Network Mobile 74

All Things Distributed

NOVEMBER 19, 2010

There was an excellent first benchmarking report of the Cluster GPU Instances by the folks at Cycle Computing - " A Couple More Nails in the Coffin of the Private Compute Cluster " The Top500 supercomputer list. Driving Storage Costs Down for AWS Customers. Expanding the Cloud - The AWS Storage Gateway. At werner.ly

AWS 64

AWS Cloud Benchmarking Storage 64

Alex Russell

APRIL 29, 2021

As an engineer on a browser team, I'm privy to the blow-by-blow of various performance projects, benchmark fire drills, and the ways performance marketing (deeply) impacts engineering priorities. With each team, benchmarks lost are understood as bugs. Chrome has missed several APIs for 3+ years: Storage Access API.

Media 145

Media Games Education Engineering 145

HammerDB

OCTOBER 12, 2018

As is also the case this limitation is at the database level (especially the storage engine) rather than the hardware level. InnoDB is the storage engine that will deliver the best OLTP throughput and should be chosen for this test. . maximum transition latency: Cannot determine or is not supported. .

Best Practices 40

Best Practices Scalability Performance C++ 40

SQL Server According to Bob

JANUARY 11, 2020

Stable Media Stable media is often confused with physical storage. SQL Server defines stable media as storage that can survive system restart or common failure. Stable media is commonly physical disk storage, but other devices and certain caching facilities qualify as well. See the article for more details. SQL Server 7.0

Servers 40

Servers Cache Media Hardware 40

ScaleOut Software

MAY 15, 2014

For example, the IMDG must be able to efficiently create millions of objects in each server to make use of its huge storage capacity. Likewise, object access paths must be heavily multi-threaded and avoid lock contention to minimize access latency and maximize throughput. Testing Scale-Up Performance.

Cache 40

Cache Servers Network Design 40

SQL Server According to Bob

JANUARY 11, 2020

​​ The storage ​​ space that is required ​​ for the sparse file is only that of the actual bytes written to the file and not the ​​ maximum ​​ file ​​ size.

Servers 40

Servers Cache Database Media 40

Smashing Magazine

JANUARY 6, 2020

Estimated Input Latency tells us if we are hitting that threshold, and ideally, it should be below 50ms. Designed for the modern web, it responds to actual congestion, rather than packet loss like TCP does, it is significantly faster , with higher throughput and lower latency — and the algorithm works differently.

Performance 145

Performance Cache Network Servers 145

Smashing Magazine

JANUARY 11, 2021

Estimated Input Latency tells us if we are hitting that threshold, and ideally, it should be below 50ms. Designed for the modern web, it responds to actual congestion, rather than packet loss like TCP does, it is significantly faster , with higher throughput and lower latency — and the algorithm works differently.

Performance 145

Performance Cache Media Metrics 145