The Netflix TechBlog

JANUARY 24, 2022

Technically, “performance” metrics are those relating to the responsiveness or latency of the app, including start up time. Every test runs on a combination of devices (physical and virtual) and platform versions ( SDKs ). and often before they are even committed to the codebase. What do we mean by Performance?

Performance 208

Performance Performance Testing Metrics Testing 208

VoltDB

OCTOBER 30, 2024

Here’s a quick comparison: Preventive maintenance: Planned in advance, cost-effective, reduces downtime, and improves reliability. Digital twins : Creates virtual models of physical assets for real-time analysis, offering insights into asset performance and maintenance needs.

Automotive 52

Automotive IoT Artificial Intelligence Energy 52

Percona

APRIL 7, 2023

Here’s how the same test performed when running Percona Distribution for PostgreSQL 14 on these same servers: Queries: reads Queries: writes Queries: other Queries: total Transactions Latency (95th) MySQL (A) 1584986 1645000 245322 3475308 122277 20137.61 We have long been surfing the virtualization wave (to keep it broad).

Tuning 145

Tuning Cache Servers Benchmarking 145

Brendan Gregg

OCTOBER 8, 2018

more capable, and built from the ground up for the modern era of the eBPF virtual machine. eBPF was created by Alexei Starovoitov while at PLUMgrid (he's now at Facebook) as a generic in-kernel virtual machine, with software defined networks as the primary use case. It's shaping up to be a DTrace version 2.0: eBPF does more.

C++ 105

C++ Virtualization Programming Latency 105

Brendan Gregg

AUGUST 8, 2017

rc5-virtual (bgregg-xenial-bpf-i-0b7296777a2585be1) 08/01/2017 _x86_64_ (8 CPU) 10:15:51 PM UID PID %usr %system %guest %CPU CPU Command 10:16:51 PM 0 18468 2.85 Some people have found values that seem to work for their systems and workloads: they know that when load goes over X, application latency is high and customers start complaining.

Latency 111

Latency Metrics C++ Systems 111

HammerDB

JULY 6, 2024

This is a brief post to highlight the metrics to use to do the comparison using a separate hardware platform for illustration purposes. Throughput: events/s (eps): 8162.5668 time elapsed: 300.0356s total number of events: 2449061 Latency (ms): min: 0.35 To run the workload on sysbench-tpcc is the following. 95th percentile: 41.10

C++ 52

C++ Hardware Benchmarking Virtualization 52

Alex Russell

MAY 9, 2022

This is a complex topic, but to borrow from a recent post , web performance expands access to information and services by reducing latency and variance across interactions in a session, with a particular focus on the tail of the distribution (P75+). Consistent performance matters just as much as low average latency.

Performance 99

Performance Latency Metrics Engineering 99

Brendan Gregg

SEPTEMBER 25, 2021

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. CLI tools The Cassandra systems were EC2 virtual machine (Xen) instances. Note that Ubuntu also has a frame to show entry into vDSO (virtual dynamic shared object). us on Centos and 0.68 us on Ubuntu. Ubuntu is 5x slower.

Speed 52

Speed Java AWS Virtualization 52

The Morning Paper

DECEMBER 8, 2019

Therefore any programming abstraction must be low latency and the kernel needs to be kept off the path of persistent data access as much as possible. Traditional pointers address a memory location (often virtual of course). This means that the overheads of system calls become much more noticeable.

Hardware 56

Hardware Programming Media Storage 56

Brendan Gregg

SEPTEMBER 5, 2017

Here's some output from my zfsdist tool, in bcc/BPF, which measures ZFS latency as a histogram on Linux: # zfsdist. Tracing ZFS operation latency. Oracle have a similar useful page as well: the Linux to Oracle Solaris 11 comparison, as well as a [procedure] for migrating from Solaris to Linux. Hit Ctrl-C to end. ^C

Virtualization 75

Virtualization AWS Engineering Hardware 75

IO River

NOVEMBER 2, 2023

This type of traffic originates directly from the server, making it more challenging to handle due to latency and server load considerations; it’s hard but not impossible. Statistics reveal that a 1% improvement in latency can lead to a 3% increase in viewer engagement, highlighting its significance in live content delivery.3.

Architecture 52

Architecture Performance Internet Internet 52

IO River

NOVEMBER 2, 2023

This type of traffic originates directly from the server, making it more challenging to handle due to latency and server load considerations; it’s hard but not impossible. This is mainly because video files are notably bulkier than a simple text-based API request, which, in comparison, would be more like a droplet than an entire river.

Architecture 52

Architecture Performance Internet Internet 52

HammerDB

OCTOBER 23, 2023

Stored Procedures and Client SQL comparison To test the stored procedures and client implementations, we ran both workloads against a system equipped with Intel Xeon 8280L. The data shows a scripted automated workload running a number of back to back tests each time with an increasing number of virtual users. On MySQL, we saw a 1.5X

Benchmarking 63

Benchmarking Database Network C++ 63

Brendan Gregg

SEPTEMBER 25, 2021

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. CLI tools The Cassandra systems were EC2 virtual machine (Xen) instances. Note that Ubuntu also has a frame to show entry into vDSO (virtual dynamic shared object). us on Centos and 0.68 us on Ubuntu. Ubuntu is 5x slower.

Speed 40

Speed Java AWS Virtualization 40

Alex Russell

APRIL 29, 2021

Now in development in WebKit after years of radio silence, WebXR APIs provide Augmented Reality and Virtual Reality input and scene information to web applications. For heavily latency-sensitive use-cases like WebXR, this is a critical component in delivering a good experience. Offscreen Canvas.

Media 145

Media Games Education Engineering 145

ScaleOut Software

MAY 15, 2014

Likewise, object access paths must be heavily multi-threaded and avoid lock contention to minimize access latency and maximize throughput. These are areas in which we have invested heavily to take advantage of 10 Gbps (and faster) networks and to handle intermittent network delays inherent in virtual server infrastructures.

Cache 40

Cache Servers Network Design 40

Brendan Gregg

OCTOBER 7, 2018

more capable, and built from the ground up for the modern era of the eBPF virtual machine. eBPF was created by Alexei Starovoitov while at PLUMgrid (he's now at Facebook) as a generic in-kernel virtual machine, with software defined networks as the primary use case. It's shaping up to be a DTrace version 2.0: eBPF does more.

C++ 40

C++ Virtualization Programming Latency 40

HammerDB

OCTOBER 12, 2018

Finally it is also important to note that this comparison is focused around OLTP based workloads, HammerDB also supports a TPC-H based workload for analytics with complex ad-hoc queries. maximum transition latency: Cannot determine or is not supported. . Error in Virtual User 1: mysqlexec/db server: Table 'mysql.proc' doesn't exist.

Best Practices 40

Best Practices Scalability Performance C++ 40

Scalegrid

JULY 29, 2020

In our final post, we will put them head-to-head in a detailed feature comparison and compare the results of PgBouncer vs. Pgpool-II performance for your PostgreSQL hosting ! PgBouncer provides a virtual database that reports various useful statistics. PostgreSQL Connection Pooling Series. Part 1 – Pros & Cons.

Benchmarking 246

Benchmarking Database Performance Testing Availability 246

SQL Server According to Bob

JANUARY 11, 2020

The caching of data pages and grouping of log records helps remove much, if not all, of the command latency associated with a write operation. ​​ The following table outlines the virtual protection states. Page State Virtual Protection State Dirty Read Write ​​ during the modification.

Servers 40

Servers Cache Database Media 40

The Netflix TechBlog

OCTOBER 14, 2024

Using this approach, we observed latencies ranging from 1 to 10 seconds, averaging 7.4 To efficiently utilize our compute resources, Titus employs a CPU oversubscription feature , meaning the combined virtual CPUs allocated to containers exceed the number of available physical CPUs on a Titus agent. We then exported the .har

Latency 214

Latency Virtualization Traffic Processing 214

Smashing Magazine

JANUARY 11, 2021

Alternatively, you can also use: Addy Osmani’s Chrome UX Report Compare Tool , Speed Scorecard (also provides a revenue impact estimator), Real User Experience Test Comparison or SiteSpeed CI (based on synthetic testing). Estimated Input Latency tells us if we are hitting that threshold, and ideally, it should be below 50ms.

Performance 145

Performance Cache Metrics Media 145