John McCalpin

FEBRUARY 17, 2025

The Multicore Era Over the past ~15 years, server processors from Intel and AMD have evolved from the early quad-core processors to the current monsters with over 50 cores per socket. The example below is for a 2005-era processor with 60 ns memory latency and 6.4 If we want to sustain full bandwidth, we need 64/2 =32 cache lines.

Latency 68

Latency Hardware Cache Systems 68

The Morning Paper

NOVEMBER 15, 2020

Achieving 100 Gbps intrusion prevention on a single server , Zhao et al., Today’s paper choice is a wonderful example of pushing the state of the art on a single server. This makes the whole system latency sensitive. Moreover, Pigasus wants to do all this on a single server! Can you really do all this on a single server??

Servers 128

Servers Hardware Latency Design 128

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 241

Cloud Traffic Best Practices Strategy 241

Scalegrid

FEBRUARY 6, 2025

Its partitioned log architecture supports both queuing and publish-subscribe models, allowing it to handle large-scale event processing with minimal latency. Kafka clusters can be deployed in Kubernetes using Helm charts to simplify scaling and management across multiple servers.

Latency 147

Latency Analytics Architecture Storage 147

VoltDB

MAY 14, 2024

Balancing Low Latency, High Availability and Cloud Choice Cloud hosting is no longer just an option — it’s now, in many cases, the default choice. As a result, IT teams picked hardware somewhat blindly but with a strong bias towards oversizing for the sake of expanding the budget, leading to systems running at 10-15% of maximum capacity.

Latency 52

Latency Availability Cloud Hardware 52

All Things Distributed

JANUARY 18, 2012

Behind the scenes, Amazon DynamoDB automatically spreads the data and traffic for a table over a sufficient number of servers to meet the request capacity specified by the customer. Amazon DynamoDB offers low, predictable latencies at any scale. s read latency, particularly as dataset sizes grow. Consistency. SimpleDBâ??s

Scalability 162

Scalability Database Ecommerce Latency 162

Scalegrid

JANUARY 25, 2024

You will need to know which monitoring metrics for Redis to watch and a tool to monitor these critical server metrics to ensure its health. Understanding Redis Performance Indicators Redis is designed to handle high traffic and low latency with its in-memory data store and efficient data structures.

Metrics 130

Metrics Monitoring Latency Cache 130

Scalegrid

JANUARY 21, 2025

Identifying key Redis metrics such as latency, CPU usage, and memory metrics is crucial for effective Redis monitoring. To monitor Redis instances effectively, collect Redis metrics focusing on cache hit ratio, memory allocated, and latency threshold.

Strategy 130

Strategy Monitoring Latency DevOps 130

Brendan Gregg

JULY 4, 2021

I summarized these topics and more as a plenary conference talk, including my own predictions (as a senior performance engineer) for the future of computing performance, with a focus on back-end servers. This was a chance to talk about other things I've been working on, such as the present and future of hardware performance.

Performance 121

Performance Latency Hardware Storage 121

Percona

MARCH 23, 2023

Hardware Memory The amount of RAM to be provisioned for database servers can vary greatly depending on the size of the database and the specific requirements of the company. Some servers may need a few GBs of RAM, while others may need hundreds of GBs or even terabytes of RAM. Benchmark before you decide.

Performance 138

Performance Hardware Tuning Storage 138

High Scalability

SEPTEMBER 14, 2018

72 : signals sensed from a distant galaxy using AI; 12M : reddit posts per month; 10 trillion : per day Google generated test inputs with 100s of servers for several months using OSS-Fuzz; 200% : growth in Cloud Native technologies used in production; $13 trillion : potential economic impact of AI by 2030; 1.8 They'll love you even more.

Internet 30

Internet Internet Scalability Education 30

The Morning Paper

OCTOBER 4, 2020

We are standing on the eve of the 5G era… 5G, as a monumental shift in cellular communication technology, holds tremendous potential for spurring innovations across many vertical industries, with its promised multi-Gbps speed, sub-10 ms low latency, and massive connectivity. Throughput and latency. energy consumption).

Energy 130

Energy Latency Performance Network 130

Scalegrid

APRIL 16, 2020

This is why our BYOC pricing is less than our Dedicated Hosting pricing, as the costs listed for BYOC are only what you pay for ScaleGrid and don’t include your hardware costs. Deploying your application and database on the same VPC also provides the lowest possible latency path. Where to host your cloud database? Security Groups.

Cloud 242

Cloud Azure AWS Database 242

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., The paper examines the implications of microservices at the hardware, OS and networking stack, cluster management, and application framework levels, as well as the impact of tail latency.

Open Source 61

Open Source Hardware Benchmarking Systems 61

Dynatrace

AUGUST 5, 2019

Lift & Shift is where you basically just move physical or virtual hosts to the cloud – essentially you just run your host on somebody else’s hardware. Remember: This is a critical aspect as you do not want to migrate a service and suddenly introduce high latency or costs to a system that you forgot about having a dependency with!

Cloud 38

Cloud Traffic Database Network 38

All Things Distributed

MARCH 11, 2016

This is a given, whether you are using the highest quality hardware or lowest cost components. When customers left the constraining, old world of IT hardware and datacenters behind, they started to develop systems with new and interesting usage patterns that no one had ever seen before. Primitives not frameworks. APIs are forever.

AWS 127

AWS Hardware Retail Virtualization 127

Scalegrid

DECEMBER 21, 2023

Identifying key Redis® metrics such as latency, CPU usage, and memory metrics is crucial for effective Redis monitoring. To monitor Redis® instances effectively, collect Redis metrics focusing on cache hit ratio, memory allocated, and latency threshold.

Strategy 130

Strategy Monitoring Latency DevOps 130

Percona

MAY 1, 2023

After some time of receiving these messages, eventually, they hit performance issues to the point that the server becomes unresponsive for a few minutes. The innodb_io_capacity_max parameter was set to 2000, so the hardware should be able to deliver that many IOPS without major issues. After that, things went back to normal.

AWS 131

AWS Hardware Storage Tuning 131

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

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. The kind of edge server envisaged here might, for example, be integrated with your WiFi access point. As such, web workers are a natural target to offload to a more powerful server.

Mobile 104

Mobile Cloud Latency Games 104

Highly Scalable

SEPTEMBER 18, 2012

Historically, NoSQL paid a lot of attention to tradeoffs between consistency, fault-tolerance and performance to serve geographically distributed systems, low-latency or highly available applications. A database should accommodate itself to different data distributions, cluster topologies and hardware configurations. Data Placement.

Database 213

Database Latency C++ Scalability 213

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 Steps Provisioning The first step is to provision the four nodes with both PostgreSQL and Citus. psql pgbench <<_eof1_ qecho adding node citus3.

Database 112

Database Benchmarking Latency C++ 112

All Things Distributed

NOVEMBER 15, 2010

For example, the most fundamental abstraction trade-off has always been latency versus throughput. Modern CPUs strongly favor lower latency of operations with clock cycles in the nanoseconds and we have built general purpose software architectures that can exploit these low latencies very well. General Purpose GPU programming.

AWS 125

AWS Programming Latency Architecture 125

The Netflix TechBlog

FEBRUARY 9, 2021

Server-generated assets, since client-side generation would require the retrieval of many individual images, which would increase latency and time-to-render. To reduce latency, assets should be generated in an offline fashion and not in real time. Localized images for each of the titles.

Engineering 168

Engineering Storage Entertainment Latency 168

Scalegrid

FEBRUARY 8, 2024

A distributed storage system is foundational in today’s data-driven landscape, ensuring data spread over multiple servers is reliable, accessible, and manageable. These storage nodes collaborate to manage and disseminate the data across numerous servers spanning multiple data centers.

Storage 130

Storage Systems Big Data Azure 130

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. " Entry/exit in/out of V8 contexts is less expensive than hardware-based isolation mechanisms, keeping request processing latency low and throughput high.

Serverless 99

Serverless Storage Latency Cloud 99

All Things Distributed

JULY 13, 2010

In particular this has been true for applications based on algorithms - often MPI-based - that depend on frequent low-latency communication and/or require significant cross sectional bandwidth. There is no more need for hardware tinkering to keep the clusters up and running (I spent many nights doing this; there is no glory in it).

Cloud 98

Cloud AWS Automotive Latency 98

Percona

MAY 2, 2023

Questions Q: I have a MySQL server with 500 GB of RAM; my data set is 100 GB. 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. Q: I have a MySQL server, and my application is writing at a rate of 100 MB/hour in my redo logs.

Social Media 134

Social Media Database Servers Cache 134

Percona

JUNE 22, 2023

As a MySQL database administrator, keeping a close eye on the performance of your MySQL server is crucial to ensure optimal database operations. However, simply deploying a monitoring tool is not enough; you need to know which Key Performance Indicators (KPIs) to monitor to gain insights into your MySQL server’s health and performance.

Performance 117

Performance Monitoring Traffic Database 117

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. Ford, et al., “TCP

Performance 103

Performance Latency Cache Virtualization 103

All Things Distributed

JANUARY 23, 2012

VM Import allows our customers to move virtual machine images from their datacenters to the Cloud and Amazon Direct Connect makes the network latencies and bandwidth between on-premises and AWS more predictable. AWS Identity and Access Management brings together on-premises and cloud identity management.

Storage 96

Storage AWS Virtualization Cloud 96

Dotcom-Montior

MAY 12, 2020

Early web applications involved less on client-side behavior and more server-side for all its navigation, query handling, and updates. A request will be sent from the client-side and an HTTP check waits on the server port to get the message, process it, and then send back the response. Connection closed by the server.

Monitoring 63

Monitoring Entertainment Hardware Traffic 63

Dotcom-Montior

NOVEMBER 24, 2021

Do you have a web server? Is the web server running? The last item to check was if the web server was able to talk to the database? These systems can include physical servers, containers, virtual machines, or even a device, or node, that connects and communicates with the network. Do you have a database? Peer-to-Peer.

Systems 74

Systems Monitoring Hardware Network 74

Brendan Gregg

AUGUST 8, 2017

Thankfully, I found some older linux-devel mailing list archives, rescued from server backups, often stored as tarballs of digests. It's time in some cgroup paths, but this server is not doing much disk I/O. Latency was acceptable and no one complained. My search was starting to feel cursed. to the load average. Yes, I'd say so.

Latency 111

Latency Metrics C++ Systems 111

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. First off there still is a model of course (but then there are servers hiding behind a serverless abstraction too!). autoscaling).

Hardware 58

Hardware Latency Serverless Energy 58

VoltDB

JULY 24, 2024

The raw performance improvement is due to c7g not using hardware threading, while Intel uses two hardware threads per physical core, which increases latency once you get beyond 50% CPU usage. File systems and IO Regardless of size, each server always has the same IO configuration. We then shifted to a bigger server.

AWS 52

AWS Benchmarking Hardware Latency 52

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