Brendan Gregg

FEBRUARY 28, 2023

My personal opinion is that I don't see a widespread need for more capacity given horizontal scaling and servers that can already exceed 1 Tbyte of DRAM; bandwidth is also helpful, but I'd be concerned about the increased latency for adding a hop to more memory.

Performance 103

Performance Latency Cache Virtualization 103

Brendan Gregg

NOVEMBER 29, 2017

In this configuration, the AMI and boot is paravirt (PV), the kernel is making hypercalls instead of privileged instructions, and the system is using paravirt network and storage drivers. But not all workloads: some are network bound (proxies) and storage bound (databases). ## 5. The AMI and boot are now HVM.

Virtualization 110

Virtualization AWS Hardware Storage 110

Percona

SEPTEMBER 19, 2023

This issue most often occurs when there is significant MySQL server activity during the backup operation, and the redo log file storage media operates faster than the backup storage media. Enabling redo log archiving on the server requires setting a value for the innodb_redo_log_archive_dirs system variable.

C++ 112

C++ Servers Storage Speed 112

Scalegrid

JANUARY 19, 2024

Recovering from a MongoDB rollback involves locating rollback files, using mongorestore to load data into a separate server, cleansing unneeded data, and importing the data back into the primary cluster, coupled with ongoing monitoring and maintenance of replica sets to maintain system health.

Database 130

Database Network Servers Monitoring 130

SQL Server According to Bob

NOVEMBER 8, 2016

SQL Server 2016 Service Pack 1 (all SKUs) , in combination with Windows Server 2016 (All SKUs) or Windows 10 Client introduces non-volatile memory support for the tail of the log file (LDF) which can significantly increase transaction throughput. Block storage is what you think of today as disk access.

Cache 40

Cache Servers Media Storage 40

Alex Russell

MARCH 6, 2021

Back in 2016, I gave a talk outlining the causes and effects of the terrible performance of web apps built using popular tools on the fastest-growing device segment: low-end to mid-range Android phones. A then-representative $200USD device had 4-8 slow (in-order, low-cache) cores, ~2GiB of RAM, and relatively slow MLC NAND flash storage.

Performance 127

Performance Network Metrics Cache 127

Percona

JANUARY 24, 2023

Some of the built-in features ( wal_compression ) have been there since 2016, and almost all backup tools do the WAL compression before taking it to the backup repository. Zstd can be chosen if the server load is not CPU bound because it can give us better compression at the expense of more CPU utilization.

Database 121

Database Benchmarking Open Source Latency 121

Percona

AUGUST 18, 2023

Why do you need multiple instances on the same server? You can have a host with two or three instances configured as a delayed replica of the source server with SQL Delay of, let’s say, 24hr, 12hr, and 6/3hrs. You can run multiple instances on a server to test your backups with the correct version and configs.

Programming 89

Programming Servers Database Systems 89

SQL Performance

MAY 1, 2020

In a recent tip , I described a scenario where a SQL Server 2016 instance seemed to be struggling with checkpoint times. I was a bit perplexed by this issue, since the system was certainly no slouch — plenty of cores, 3TB of memory, and XtremIO storage. SQL 2016 – It Just Runs Faster: Indirect Checkpoint Default.

C++ 58

C++ Database Servers Storage 58

SQL Performance

SEPTEMBER 5, 2019

I briefly mentioned that batch mode data is normalized in my last article Batch Mode Bitmaps in SQL Server. You will need SQL Server 2016 (or later) and Developer Edition (or equivalent) to reproduce the results shown here. This table uses columnstore for its primary storage to produce batch mode execution later on.

Performance 81

Performance C++ Storage Servers 81

Brendan Gregg

JULY 4, 2021

AWS Graviton2); for memory with the arrival of DDR5 and High Bandwidth Memory (HBM) on-processor; for storage including new uses for 3D Xpoint as a 3D NAND accelerator; for networking with the rise of QUIC and eXpress Data Path (XDP); and so on.

Performance 52

Performance Latency Hardware Storage 52

SQL Performance

MAY 29, 2019

It can be activated from SQL Server 2008 to 2014 inclusive using documented trace flag 610. From SQL Server 2016 onward, FastLoadContext is enabled by default ; the trace flag is not required. For more background, see the Data Performance Loading Guide and the Tiger Team notes on the behaviour changes for SQL Server 2016.

Cache 64

Cache Testing Servers Code 64

SQL Performance

APRIL 23, 2019

In SQL Server 2012, grouped (vector) aggregation was able to use parallel batch-mode execution, but only for the partial (per-thread) aggregate. SQL Server 2014 added the ability to perform parallel batch-mode grouped aggregation within a single Hash Match Aggregate operator. Columnstore storage. Introduction.

Servers 67

Servers Testing Metrics Storage 67

Brendan Gregg

SEPTEMBER 5, 2017

LTS (April 2016). I wrote about it in a previous post, [DTrace for Linux 2016]. I wrote a page on it: [perf]. - **eBPF**: tracing features completed in 2016, this provides efficient programmatic tracing to existing kernel frameworks. Oracle VM Server is based on Xen. The hardest part on Linux is now done: kernel support.

Virtualization 75

Virtualization AWS Engineering Hardware 75

Percona

JANUARY 17, 2019

now has a version which will support parallelism for SELECT queries (utilizing the read capacity of storage nodes underneath the Aurora cluster). I will compare AWS Aurora with MySQL (Percona Server) 5.6 84.1 | | version_comment | Percona Server (GPL), Release 84.1, MySQL on ec2.

Cache 48

Cache C++ AWS Airlines 48

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. These AMD EPYC processors have a number of advantages for SQL Server workloads, as I will explain in this article. Since then, Microsoft has changed the naming of this series to Lsv2.

Azure 42

Azure Virtualization Storage Benchmarking 42

Brendan Gregg

FEBRUARY 28, 2023

My personal opinion is that I don't see a widespread need for more capacity given horizontal scaling and servers that can already exceed 1 Tbyte of DRAM; bandwidth is also helpful, but I'd be concerned about the increased latency for adding a hop to more memory.

Performance 52

Performance Latency Cache Virtualization 52

Percona

JUNE 4, 2023

In this blog, we’ll provide a comparison between MariaDB vs. MySQL (including Percona Server for MySQL ). Introduction: MariaDB vs. MySQL The goal of this blog post is to evaluate, at a higher level, MariaDB vs. MySQL vs. Percona Server for MySQL side-by-side to better inform the decision making process.

Open Source 52

Open Source Servers Storage Scalability 52

SQL Performance

MAY 16, 2019

Without this guarantee, SQL Server might add index rows that are not sorted correctly, which would not be good. For INSERT.SELECT , SQL Server makes its own determination whether to ensure rows are presented to the Clustered Index Insert operator in key order or not. DMLRequestSort Conditions.

Cache 67

Cache Servers Testing Storage 67

SQL Performance

MAY 1, 2019

The summary rows for heap tables without indexes are the same in both documents (no changes for SQL Server 2016): An explicit TABLOCK hint is not the only way to meet the requirement for table-level locking. The inputs to this calculation are: The version of SQL Server. Target table row size. The Data Size Calculation.

Servers 47

Servers Database Testing Performance 47

Smashing Magazine

JANUARY 15, 2019

So, apart from powering servers with renewable energy, what else can web developers do about climate change? This includes the work done by the server, the client and the intermediary communications networks that transmit data between the two. In 2016, O’Reilly published “ Designing For Sustainability ” by Tim Frick.

Website Performance 74

Website Performance Website Energy Performance 74

Alex Russell

OCTOBER 22, 2017

Late-loading JavaScript can cause “server-side rendered” pages to fail in infuriating ways. The server sends it as a stream of bytes and when the browser encounters each of the sub-resources referenced in the document, it requests them. The true median device from 2016 sold at about ~$200 unlocked.

Performance 74

Performance Network Benchmarking Mobile 74

Brendan Gregg

NOVEMBER 28, 2017

In this configuration, the AMI and boot is paravirt (PV), the kernel is making hypercalls instead of privileged instructions, and the system is using paravirt network and storage drivers. But not all workloads: some are network bound (proxies) and storage bound (databases). ## 5. The AMI and boot are now HVM.

Virtualization 40

Virtualization AWS Hardware Storage 40

CSS Wizardry

MARCH 3, 2019

This includes CDNs, proxy servers, and the like. Once the 60 seconds is up, the browser will head back to the server to revalidate the file. It means ‘do no t serve a copy from cache until you’ve revalidated it with the server and the server said you can use the cached copy’. public and private. Cache-Control: no-cache.

Cache 266

Cache Latency Strategy Servers 266

SQL Server According to Bob

DECEMBER 18, 2018

SQL Server relies on Forced-Unit-Access (Fua) I/O subsystem capabilities to provide data durability, detailed in the following documents: SQL Server 2000 I/O Basic and SQL Server I/O Basics, Chapter 2. Be sure to deploy SQL Server 2017 CU6 or newer for best data durability and performance results. “. 4 Socket, TPCC.

Servers 90

Servers Media Cache Storage 90

Percona

JANUARY 14, 2019

In a partitioned massively parallel database system, the storage format and sorting algorithm may not be optimized for that operation as we are reading multiple partitions in parallel. voting','maven'] ? ? compose','xamarin','markdown','scrum','comic'] ? ? china','sourceforge','subscription','chinese','kotlin'] ? ?

Database 40

Database Analytics Blockchain Healthcare 40