HammerDB

OCTOBER 18, 2018

An essential part of database performance testing is viewing the statistics generated by the database during the test and in 2009 HammerDB introduced automatic AWR snapshot generation for Oracle for the TPC-C test. With this feature Oracle generates a wealth of performance data that can be reviewed once the test is complete.

Benchmarking 40

Benchmarking Monitoring Performance C++ 40

Adrian Cockcroft

JUNE 21, 2022

Another big jump, but now it was my job to run benchmarks in the lab, and write white papers that explained the new products to the world, as they were launched. I was mostly coding in C, tuning FORTRAN, and when I needed to do a lot of data analysis of benchmark results used the S-PLUS statistics language, that is the predecessor to R.

Tuning 52

Tuning Benchmarking Engineering C++ 52

HammerDB

FEBRUARY 25, 2021

HammerDB is a load testing and benchmarking application for relational databases. However, it is crucial that the benchmarking application does not have inherent bottlenecks that artificially limits the scalability of the database. Basic Benchmarking Concepts. To benchmark a database we introduce the concept of a Virtual User.

Programming 67

Programming Benchmarking Virtualization C++ 67

Brendan Gregg

SEPTEMBER 25, 2021

As (C) looked like a kernel rebuild, I started with (D) and (E). ## 5. I also rewrote this in C and called gettimeofday(2) directly: $ cat gettimeofdaybench.c. Checking those available: $ cat /sys/devices/system/clocksource/clocksource0/available_clocksource. Microbenchmark os::javaTimeMillis() on both systems.

Speed 126

Speed Java AWS Virtualization 126

Brendan Gregg

FEBRUARY 9, 2018

In this post I'll look at the Linux kernel page table isolation (KPTI) patches that workaround Meltdown: what overheads to expect, and ways to tune them. I then analyzed performance during the benchmark ([active benchmarking]), and used other benchmarks to confirm findings. Hit Ctrl-C to end. and 4.14.12

Performance 75

Performance Benchmarking Cache Tuning 75

Brendan Gregg

SEPTEMBER 25, 2021

As (C) looked like a kernel rebuild, I started with (D) and (E). ## 6. I also rewrote this in C and called gettimeofday(2) directly: $ cat gettimeofdaybench.c Here's some experimental approaches I could also explore: Disable tracesys/syscall_trace. Microbenchmark os::javaTimeMillis() on both systems. include <sys/time.h>

Speed 40

Speed Java AWS Virtualization 40

HammerDB

OCTOBER 12, 2018

This post complements the previous best practice guides this time with the focus on MySQL and MariaDB and achieving top levels of performance with the HammerDB MySQL TPC-C test. SELECT DISTINCT c from sbtest where id between ? order by c. SELECT c from sbtest where id=? HammerDB difference from Sysbench. perf special.

Best Practices 40

Best Practices Scalability Performance C++ 40

Brendan Gregg

FEBRUARY 8, 2018

In this post I'll look at the Linux kernel page table isolation (KPTI) patches that workaround Meltdown: what overheads to expect, and ways to tune them. I then analyzed performance during the benchmark ([active benchmarking]), and used other benchmarks to confirm findings. Hit Ctrl-C to end. and 4.14.12

Performance 40

Performance Benchmarking Cache Tuning 40

Randon ASCII

APRIL 17, 2018

When you own all of the code then this may involve some back of the envelope estimates, competitive benchmarking, or intuition tuned by experience. The diagnosis is that file notification buffers are allocated out of pool memory ( ExAllocatePoolWithTag ) and then filled with the available notification data.

Monitoring 43

Monitoring C++ Code Processing 43

John McCalpin

JANUARY 22, 2018

Since “peak performance” for the processor is two (512-bit SIMD) FMA instructions per cycle, any instructions that are not FMA instructions subtract directly from the available peak performance. of the “adjusted peak performance”, there is no longer a significant upside to performance tuning.

Latency 40

Latency Hardware Code Testing 40

High Scalability

NOVEMBER 25, 2019

Always be available: Have a service level circuit breaker and a circuit breaker by customer. We changed deployment in the middle of the week for both work-life balance and also this way we would have all engineers available in case release runs into an issue. Java/Python mostly and some small services in Go/C. Cloud Platform.

Architecture 52

Architecture Cache Azure Storage 52

SQL Server According to Bob

JANUARY 11, 2020

​​ Various defragmentation ​​ utilities ​​ are ​​ available ​​ on the market today. Therefore, the database can have pages in any one of the three available verification states.

Servers 40

Servers Cache Database Media 40