Percona

NOVEMBER 3, 2023

PostgreSQL Cluster One coordinator node citus-coord-01 Three worker nodes citus1 citus2 citus3 Hardware AWS Instance Ubuntu Server 20.04, SSD volume type 64-bit (x86) c5.xlarge Repeating the benchmark once again demonstrates that performance has degraded somewhat. NOTICE: Moving shard 102022 from citus1:5432 to citus2:5432.

Database 109

Database Benchmarking Latency C++ 109

Dynatrace

APRIL 5, 2021

Real-time stream processing to perform live activity tracking, data cleansing, metrics generation, and more. The function itself performs a small unit of work and Lambda charges subscribers by the millisecond. For such functions, organizations will be better off using an EC2 instance or their own hardware to interact with data.

Lambda 180

Lambda AWS Serverless Hardware 180

Randon ASCII

JANUARY 12, 2022

That meant I started having regular meetings with the hardware engineers who were working with IBM on the CPU which gave me even more expertise on this CPU, which was critical in helping me discover a design flaw in one of its instructions , and in helping game developers master this finicky beast. So, anyway. Standard stuff.

Cache 126

Cache Latency Benchmarking Hardware 126

The Morning Paper

MAY 12, 2019

An open-source benchmark suite for microservices and their hardware-software implications for cloud & edge systems Gan et al., Microservices fundamentally change a lot of assumptions current cloud systems are designed with, and present both opportunities and challenges when optimizing for quality of service (QoS) and utilization.

Open Source 61

Open Source Hardware Benchmarking Systems 61

The Morning Paper

NOVEMBER 15, 2020

With more nodes and more coordination comes more complexity, both in design and operation. This makes the whole system latency sensitive. One of the joys of this paper is that you don’t just get to see the final design, you also get insight into the forces and trade-offs that led to it. Back of the envelope.

Servers 128

Servers Hardware Latency Design 128

The Netflix TechBlog

NOVEMBER 9, 2022

A quick canary test was free of errors and showed lower latency, which is expected given that our standard canary setup routes an equal amount of traffic to both the baseline running on 4xl and the canary on 12xl. What’s worse, average latency degraded by more than 50%, with both CPU and latency patterns becoming more “choppy.”

Hardware 363

Hardware Cache Performance Latency 363

The Netflix TechBlog

JUNE 4, 2019

Because microprocessors are so fast, computer architecture design has evolved towards adding various levels of caching between compute units and the main memory, in order to hide the latency of bringing the bits to the brains. As it turns out, for the large majority of Netflix use cases, its performance is far from optimal.

Cache 251

Cache Latency Airlines Logistics 251

Alex Russell

JANUARY 30, 2024

The usual caveats also apply: Performance is a deep and nuanced domain, and much can go wrong beyond content size and composition. How sites manage resources after-load can have a big impact on perceived performance. Until and unless teams have better data about their performance, the global baseline budget should be enforced.

Performance 90

Performance Network Mobile Speed 90

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. Poor performance has a compound effect on user expectations at an ecosystem level. Live by the link, die by the link.

Performance 124

Performance Network Cache Metrics 124

Brendan Gregg

JULY 4, 2021

It's an exciting time for developments in computer performance, not just for the BPF technology (which I often [write about]) but also for processors with 3D stacking and cloud vendor CPUs (e.g., This was a chance to talk about other things I've been working on, such as the present and future of hardware performance.

Performance 52

Performance Latency Hardware Storage 52

ACM Sigarch

MAY 31, 2023

Heterogeneous and Composable Memory (HCM) offers a feasible solution for terabyte- or petabyte-scale systems, addressing the performance and efficiency demands of emerging big-data applications. CXL-based disaggregated memory provides a new opportunity with two orders of magnitude better performance than RDMA (as illustrated in Figure 2).

Latency 52

Latency Hardware Cache Architecture 52

Percona

APRIL 17, 2023

Note that the intent of tuning the settings is not exclusively about improving performance but also enhancing the high availability and resilience of the MongoDB database. The CFQ works well for many general use cases but lacks latency guarantees. The goal is to reduce memory usage without impacting query performance negatively.

Best Practices 52

Best Practices Design Tuning Database 52

The Netflix TechBlog

AUGUST 30, 2021

Complementing the hardware is the software on the RAE and in the cloud, and bridging the software on both ends is a bi-directional control plane. When a new hardware device is connected, the Local Registry detects and collects a set of information about it, such as networking information and ESN. million elements.

Latency 213

Latency Traffic Transportation Hardware 213

Percona

MAY 1, 2023

Default settings can help you get started quickly – but they can also cost you performance and a higher cloud bill at the end of the month. I’ll show you some MySQL settings to tune to get better performance, and cost savings, with AWS RDS. Want to save money on your AWS RDS bill? After that, things went back to normal.

AWS 124

AWS Hardware Storage Tuning 124

Brendan Gregg

FEBRUARY 28, 2023

At USENIX SREcon22 APAC I gave the opening keynote on the future of computer performance, rounding up the latest developments and making predictions of where I see things heading. This talk originated from my updates to Systems Performance 2nd Edition , and this was the first time I've given this talk in person!

Performance 52

Performance Latency Cache Virtualization 52

The Morning Paper

NOVEMBER 10, 2019

It’s been clear for a while that software designed explicitly for the data center environment will increasingly want/need to make different design trade-offs to e.g. general-purpose systems software that you might install on your own machines. The desire for CPU efficiency and lower latencies is easy to understand.

Network 92

Network Transportation Latency Entertainment 92

Dynatrace

SEPTEMBER 30, 2021

It requires purchasing, powering, and configuring physical hardware, training and retaining the staff capable of servicing and securing the machines, operating a data center, and so on. They need enough hardware to serve their anticipated volume and keep things running smoothly without buying too much or too little. Reduced cost.

Cloud 165

Cloud Traffic Best Practices Strategy 165

Dynatrace

OCTOBER 1, 2021

In these modern environments, every hardware, software, and cloud infrastructure component and every container, open-source tool, and microservice generates records of every activity. In a monitoring scenario, you typically preconfigure dashboards that are meant to alert you to performance issues you expect to see later.

Metrics 363

Metrics Open Source Monitoring Infrastructure 363

The Morning Paper

MAY 19, 2019

Last week we learned about the [increased tail-latency sensitivity of microservices based applications with high RPC fan-outs. Seer uses estimates of queue depths to mitigate latency spikes on the order of 10-100ms, in conjunction with a cluster manager. So what we have here is a glimpse of the limits for low-latency RPCs under load.

Latency 80

Latency Hardware Network Architecture 80

John McCalpin

AUGUST 1, 2018

Bandwidth, performance analysis has two recurring themes: How fast should this code (or “simple” variations on this code) run on this hardware? If I am analyzing (apparent) performance shortfalls, how can I distinguish between cause and effect ? The MPI runtime library. in ways that are seldom transparent.

Hardware 40

Hardware Transportation Performance Latency 40

Smashing Magazine

JANUARY 11, 2021

Front-End Performance Checklist 2021. Front-End Performance Checklist 2021. This guide has been kindly supported by our friends at LogRocket , a service that combines frontend performance monitoring , session replay, and product analytics to help you build better customer experiences. Web performance is a tricky beast, isn’t it?

Performance 145

Performance Cache Media Metrics 145

The Morning Paper

JUNE 13, 2019

As we saw with the SOAP paper last time out, even with a fixed model variant and hardware there are a lot of different ways to map a training workload over the available hardware. “ For instance, Facebook applications issue tens-of-trillions of inference queries per day with varying performance, accuracy, and cost constraints.”

Hardware 58

Hardware Latency Serverless Energy 58

The Morning Paper

MAY 14, 2019

Seer: leveraging big data to navigate the complexity of performance debugging in cloud microservices Gan et al., Finally, we show that Seer can identify application level design bugs, and provide insights on how to better architect microservices to achieve predictable performance. ASPLOS’19.

Big Data 45

Big Data Cloud Performance Hardware 45

John McCalpin

APRIL 25, 2023

The primary metric for memory bandwidth in multicore processors is that maximum sustained performance when using many cores. This post is about a secondary performance characteristic — sustained memory bandwidth for a single thread running on a single core. buffer copies for filesystem access) with a single thread.

Benchmarking 97

Benchmarking Cache Latency Tuning 97

SQL Performance

APRIL 27, 2020

Gen 5 is the primary hardware option now for most regions since Gen 4 is aging out. The business-critical tier has local SSD which provide much better IO performance with max data IOPS ranging from 8000 for a 2 vCore database up to 204,800 for an 80 vCore database. New Hardware Configuration for Provisioned Compute Tier.

Azure 55

Azure Database Serverless Hardware 55

The Morning Paper

JANUARY 30, 2020

Edge servers are the middle ground – more compute power than a mobile device, but with latency of just a few ms. These use their regression models to estimate processing time (which will depend on the hardware available, current load, etc.). Why would we want to live migrate web workers? Is the migration worth it though?

Mobile 104

Mobile Cloud Latency Games 104

The Morning Paper

JANUARY 28, 2020

Shredder is " a low-latency multi-tenant cloud store that allows small units of computation to be performed directly within storage nodes. " From an operator perspective it makes it harder to follow the classic cloud-native design in which a global storage layer is separate to compute. High performance.

Serverless 99

Serverless Storage Latency Cloud 99

The Morning Paper

MARCH 17, 2020

Based on these findings, the authors present four guidelines to get the best performance out of this memory today. The data we have collected demonstrate that many of the assumptions that researchers have made about how NVDIMMs would behave and perform are incorrect. Understanding Optane DIMM performance. Optane DIMMs are here!

Scalability 59

Scalability Latency Cache Media 59

Percona

MAY 2, 2023

Keep in mind that setting the buffer pool size too high may result in other processes on your server competing for memory, which can impact performance. However, it’s important to note that the optimal value may vary depending on your specific workload and hardware configuration. 16) and monitoring the server’s performance.

Social Media 124

Social Media Database Servers Cache 124

The Morning Paper

NOVEMBER 5, 2019

Breaking that assumption allowed Ceph to introduce a new storage backend called BlueStore with much better performance and predictability, and the ability to support the changing storage hardware landscape. But let’s take a quick look at the changing hardware landscape before we go on… The changing hardware landscape.

Storage 64

Storage Systems Hardware Efficiency 64

Smashing Magazine

JANUARY 6, 2020

Front-End Performance Checklist 2020 [PDF, Apple Pages, MS Word]. Front-End Performance Checklist 2020 [PDF, Apple Pages, MS Word]. Web performance is a tricky beast, isn’t it? How do we actually know where we stand in terms of performance, and what our performance bottlenecks exactly are? Vitaly Friedman.

Performance 145

Performance Cache Network Metrics 145

Brendan Gregg

AUGUST 18, 2019

bpftrace is a new open source tracer for Linux for analyzing production performance problems and troubleshooting software. Linux already has many performance tools, but these are often counter-based and have limited visibility. Linux already has many performance tools, but these are often counter-based and have limited visibility.

Latency 68

Latency C++ Cache Programming 68

The Morning Paper

MARCH 3, 2020

It takes you through the thinking processes and engineering practices behind the design of a key part of the control plane for AWS Elastic Block Storage (EBS): the Physalia database that stores configuration information. This work is latency critical, because volume IO is blocked until it is complete. NSDI’20. Requirements.

Database 106

Database AWS Network Design 106

Smashing Magazine

JANUARY 7, 2019

Front-End Performance Checklist 2019 [PDF, Apple Pages, MS Word]. Front-End Performance Checklist 2019 [PDF, Apple Pages, MS Word]. Web performance is a tricky beast, isn’t it? How do we actually know where we stand in terms of performance, and what our performance bottlenecks exactly are? Vitaly Friedman.

Performance 103

Performance Cache Metrics Network 103

The Morning Paper

JULY 4, 2019

This is a fascinating paper from members of Netflix’s Resilience Engineering team describing their chaos engineering initiatives: automated controlled experiments designed to verify hypotheses about how the system should behave under gray failure conditions, and to probe for and flush out any weaknesses. Safeguards.

Latency 77

Latency Engineering Metrics Traffic 77

VoltDB

JUNE 1, 2017

It was – like the hypothetical movie I describe above – more than a little bit odd, as you could leave a session discussing ever more abstract layers of virtualization and walk into one where they emphasized the critical importance of pinning a network interface to a specific VM for optimal performance.

Hardware 56

Hardware Latency Virtualization Cloud 56

The Morning Paper

MAY 21, 2019

This boils down to a single digit µs latency toleration in the tail for far memory, and in addition to security and privacy concerns, rules out remote memory solutions. Thus we’re fundamentally trading (de)-compression latency at access time for the ability to pack more data in memory. Enter zswap! min promotion rate though?

Software 78

Software Software Google Hardware 78