SQL Performance

OCTOBER 29, 2019

There are multiple methods to look at poorly performing queries in SQL Server, notably Query Store, Extended Events, and dynamic management views (DMVs). In order to use Query Store and Extended Events, you have to configure them in advance – either enabling Query Store for your database(s), or setting up an XE session and starting it.

C++ 74

C++ Database Servers Tuning 74

SQL Performance

JULY 1, 2019

The results are technically non-deterministic since SQL Server could validly return any one of the rows in each group. Nevertheless, if you run this query yourself, you are quite likely to see the same result I do: The execution plan depends on the version of SQL Server used, and does not depend on database compatibility level.

C++ 58

C++ Servers Database Performance 58

Brendan Gregg

SEPTEMBER 25, 2021

These strange questions came to the fore back in 2014 when Netflix was switching services from CentOS Linux to Ubuntu, and I helped debug several weird performance issues including one I'll describe here. A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. How would you _time_ time?

Speed 126

Speed Java AWS Virtualization 126

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). As we moved towards SQL Server 2014, the pace of hardware accelerated.

Availability 40

Availability Servers Hardware Benchmarking 40

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 following demo script should be run on a development instance in a new test database set to use the SIMPLE recovery model.

Servers 47

Servers Database Testing Performance 47

Scalegrid

AUGUST 15, 2019

In 2014, we discussed 4 of the top MongoDB GUIs: MongoVue, MongoHub, RockMongo, and Robo 3T (formerly Robomongo), and again in 2016: MongoDB Compass, Robo 3T, Studio 3T, and MongoBooster. Should have an excellent way to measure server and query performance. A good UI is an important part of the development experience. Click To Tweet.

Database 40

Database Servers Monitoring Availability 40

Brendan Gregg

NOVEMBER 29, 2017

At Netflix, we've been using these technologies as they've been made available for instance types in the AWS EC2 cloud. It's all a bit confusing, and I wrote about this in 2014: [Xen Modes]. When I joined Netflix in 2014, we had begun transitioning from PV to HVM (PVHVM). This recommendation quickly became out of date.

Virtualization 111

Virtualization AWS Hardware Storage 111

All Things Distributed

APRIL 4, 2017

The new AWS EU (Stockholm) Region will have three Availability Zones and will be ready for customers to use in 2018. As a result, we have opened 42 Availability Zones across 16 AWS Regions worldwide. In 2014 and 2015 respectively, AWS opened offices in Stockholm and Espoo, Finland. That’s 100% faster.

AWS 117

AWS Airlines Latency Games 117

Brendan Gregg

SEPTEMBER 25, 2021

These strange questions came to the fore back in 2014 when Netflix was switching services from CentOS Linux to Ubuntu, and I helped debug several weird performance issues including one I'll describe here. A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. How would you time time?

Speed 52

Speed Java AWS Virtualization 52

Brendan Gregg

SEPTEMBER 25, 2021

These strange questions came to the fore back in 2014 when Netflix was switching services from CentOS Linux to Ubuntu, and I helped debug several weird performance issues including one I'll describe here. A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. How would you _time_ time?

Speed 40

Speed Java AWS Virtualization 40

SQL Performance

APRIL 8, 2019

These VMs are not available in all regions, so you will want to check the availability in the Azure region that you are interested in using. 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

SQL Performance

MARCH 8, 2021

SQL Server 2014 SP2 and later produce runtime (“actual”) execution plans that can include elapsed time and CPU usage for each execution plan operator (see KB3170113 and this blog post by Pedro Lopes). SQL Server makes some timing adjustments in parallel plans to promote consistency, but they are not perfectly implemented.

Servers 95

Servers Tuning Architecture Processing 95

Brendan Gregg

NOVEMBER 28, 2017

At Netflix, we've been using these technologies as they've been made available for instance types in the AWS EC2 cloud. It's all a bit confusing, and I wrote about this in 2014: [Xen Modes]. When I joined Netflix in 2014, we had begun transitioning from PV to HVM (PVHVM). This recommendation quickly became out of date.

Virtualization 40

Virtualization AWS Hardware Storage 40

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. instructions are available. Introduction.

Servers 67

Servers Testing Metrics Storage 67

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

O'Reilly

JULY 25, 2023

To provide some coherence to the music, I decided to use Taylor Swift songs since her discography covers the time span of most papers that I typically read: Her main albums were released in 2006, 2008, 2010, 2012, 2014, 2017, 2019, 2020, and 2022. This choice also inspired me to call my project Swift Papers. But still, I liked how Moment.js

Programming 98

Programming Code Design Benchmarking 98

Alex Russell

MARCH 6, 2021

Devices and networks have evolved too: Alex Russell @slightlylate An update on mobile CPUs and the Performance Inequality Gap: Mid-tier Android devices (~$300) now get the single-core performance of a 2014 iPhone and the multi-core perf of a 2015 iPhone. Modern network performance and availability. Here begins our 2021 adventure.

Performance 124

Performance Network Cache Metrics 124

SQL Performance

MAY 7, 2019

Back in 2014, I started a series of blog posts here to talk about specific wait types and what they do and don’t mean. I’m going to introduce you to wait statistics and explain how critical they are for troubleshooting workload performance in SQL Server. SQL Server also keeps track of how long the thread has to wait. Scheduling.

Servers 40

Servers Azure Metrics Tuning 40

Percona

JANUARY 14, 2019

What if we use ClickHouse (which is a columnar analytical database) as our main datastore? Well, typically, an analytical database is not a replacement for a transactional or key/value datastore. Although such databases can be very efficient with counts and averages, some queries will be slow or simply non existent. Text search.

Database 40

Database Analytics Blockchain Healthcare 40