Questions tagged [performance]
Questions about server hardware and software performance, or network performance.
3,234 questions
1 vote
1 answer
196 views
Hugo Performance: Thousands of post subdirectories within public directory
I'm going to host a lot of Hugo posts as page bundles(directories) on a ZFS filesystem on a FreeBSD machine. So there's going to be 100,000 directories(page bundles consist of images and an index.md ...
1 vote
0 answers
86 views
Ubuntu VPS with Xfce GUI freezes a lot when connected via RDP
I have a VPS on Beget, where Ubuntu 22.04 with a lightweight Xfce GUI is deployed. I need several users (3-4 people) to be able to connect to the remote server from their local PCs simultaneously, ...
- 11
7 votes
3 answers
577 views
VMWare, SAN, SSD, Influence of the fill level on the performance of a virtual hard disk
We have several VM hosts with a Nimble (SSD) SAN. The SAN is connected with 2x10GB. The virtual machines are mostly Windows Server 2019. How does the size the free space of the drive storage affect ...
- 71
1 vote
0 answers
53 views
How to optimize server resources for a tourism booking website with traffic spikes?
I manage the infrastructure for a Red Sea diving website Dive More Group, and we experience heavy traffic during European holiday seasons. I'm hosting it on a VPS with Nginx + PHP 8.2 + MySQL, and ...
1 vote
0 answers
83 views
Diskspd results on local server a are ok, but they are not when going over smb
On a VMWare VM I use this commandline: diskspd.exe -c10G -w50 -b32K -F8 -r32K -o32 -W60 -d60 -Sh -L e:\Testfile The parameters for diskspd I'm using there are given by our software vendor. The ...
- 11
4 votes
1 answer
386 views
High I/O NFS writes causes system freeze
I have a very odd issue that I have not so far found a resolution for.. If I'm "streaming" content (download) from my NAS (via NFS) its fine (relatively low but consistent data).. but if I ...
- 41
0 votes
0 answers
149 views
Storage Spaces Direct - Slow disk I/O performance - Windows Server 2019
I'm having issues with the disk performance of a three node Windows Server 2019 failover cluster, using Storage Spaces Direct. Each node has 4 x 3.2TB SSDs and 16 x 2.4TB HDDs. The cluster storage is ...
- 1
1 vote
0 answers
203 views
RAID 530-8i PCIe 12Gb Adapter for Linux Based OS
I am repurposing a Lenovo ThinkSystem ST250 with RAID 530-8i PCIe 12Gb Adapter running 2 x 960GB SSD SATA 6Gb ThinkSystem drives in RAID 1 on 0 & 1. I had VMWare running on windows server on this ...
1 vote
0 answers
105 views
IIS site experiencing bad performance
We have an ASP.Net site on IIS 8.5 on Windows Server 2012 with sporadic bad performance. I had failed request tracing enabled, and looking through the log files I noticed this event is taking the most ...
- 11
-1 votes
1 answer
102 views
Raid: Old concept o SSD
Is RAID becoming an obsolete concept with the rise of SSDs, or does its role shift from performance enhancement to redundancy and storage capacity in modern systems?
0 votes
0 answers
110 views
Low performance in VPS vs Local machine with Docker
I have developed a web application using Laravel and PostgreSQL as the database. The main function of my application is to process large volumes of data from the database, with tables ranging from a ...
- 1
2 votes
0 answers
540 views
Windows Server 2022 NtfsDisableLastAccessUpdate Registry Key Not Disabling Last Access Time Updates
I'm trying to disable Last Access Time Updates on Windows Server 2022. It is a NTFS feature that records the last time file was accessed. This is needed to improve web server performance and is ...
- 237
0 votes
0 answers
213 views
Capacity estimations for Workload IOPS and Throughput requirements based on Disk IOPS and Throughput specifications
I am trying to understand if there is a correlation between "Workload IOPS and Throughput requirements" and "Disk IOPS and Throughput specifications". I stumbled upon a couple of ...
- 109
0 votes
0 answers
156 views
qemu VM super-slow if host is running heavy CUDA processes
(I originally asked this on SuperUser, but I think it is misplaced there, so I'm re-asking this here with some additional information; in any case sorry for cross-posting) TL;DR My host is used for ...
- 541
0 votes
0 answers
63 views
Increase PHP flush rate (2-3 times a second to 10+) for server-sent event
I have a server-sent event that processes about 20-30 items per second. On the client I have a progress element and span of status text that updates for each item processed. Right now it "chugs&...
- 133
-1 votes
1 answer
91 views
Write log files to different partition on same disk or different physical disk? [duplicate]
On a linux server dedicated to running one application that processes financial transaction input data files that arrive every minute using sftp, what is the recommendation for writing out log files? ...
- 15
2 votes
1 answer
371 views
Advice on Using NVMe Drives for ZFS Cache to Optimize Performance for Google PageSpeed Insights
I’m setting up a new server with the following ZFS-based storage tiers: Base Storage: RAID-Z2 with 6 x 1TB SAS drives (~4TB usable). SSD Storage: ZFS Mirror with 2 x 1TB SSDs (~1TB usable). NVMe ...
- 21
0 votes
0 answers
68 views
Setting thresholds for disk I/O metrics like wr_sec/s or rd_sec/s from sar command
We have a Red Hat Kafka servers, and the disk utilization on the Kafka disk is very high almost 100%. What we observe from the sar command is that the values of rd_sec/s and wr_sec/s are high. The ...
- 739
0 votes
0 answers
251 views
Recommended way to host multiple websites using docker
I read docker docs, and it says that best practices is each container for each app. So let's say that I want to host around 100 web hosts or more, what would be recommended way? So should I create for ...
- 299
0 votes
0 answers
18 views
How to manage hardware resources for a server with Nginx and PHP-fpm? [duplicate]
I have a Linux (Ubuntu 20.04) server with 32GB RAM, a 16-core CPU, SSD storage, and 16000 IOPS. Recently we had a campaign and a large traffic came on our website. The website threw 502 Bad Gateway. I ...
- 113
0 votes
0 answers
28 views
How should I configurate servers om Process Manager? [duplicate]
I have a server that has 32GB RAM and 16-core CPU and uses SSD storage and 16000 IOPS. The server OS is Ubunue (20.04) and uses Nginx, PHP-fpm to handle requests. I need to configure the server so ...
- 113
0 votes
0 answers
182 views
qemu Harddisk slow access (>10 seconds)
I'm trying to set up multiple Windows guests on a Linux server using KVM / qemu and I'm getting abysmal response on disk reads and writes. Windows 11 Task Manager on one guest is reporting disk ...
- 3,086
0 votes
0 answers
35 views
Prove website performance is due to the shared hosting provider server [duplicate]
Starting 11/09/2024 my website performance decreased after the hosting provider migrated the site to a cloud-based server. The issue occurs when transitioning from HTTP to HTTPS (about 1 or 2 seconds) ...
- 101
0 votes
0 answers
54 views
Two disks in mdadm but one has much higher aqu-sz, await, %util
I have two disks in an mdadm RAID1 array but (at times at least) they seem to be performing very differently from one another: 07:24:53 AM tps rkB/s wkB/s dkB/s areq-sz aqu-sz ...
- 53
1 vote
1 answer
96 views
zfs, possibly bad performance
I am looking to confirm if this behaviour is normal as I got a new server and I am testing some SQL queries and they seem to be slightly slower than on the old machine. This is freebsd 14.1 When I ...
- 23
0 votes
1 answer
516 views
Optimizing Disk Performance - IOPS/Throughput
I've got a VM used for SQL, and it's being suggested that splitting temp-DB, transaction logs, master, etc. onto separate physical disks will boost performance. This VM is in a datacenter (all flash ...
6 votes
2 answers
1k views
HPE Smart Array P408i 2GB experiencing a performance drop
I have two HP ProLiant ML 350 GEN 10 with a RAID 10 SSD Samsung disks and HPE SmartArray P408 with 2 GB cache. It has good performance: When copying large files, performance drops after two minutes......
0 votes
1 answer
191 views
Linux LUKS Encryption - On-demand or FULL in advance?
When encrypting a device using LUKS, it seems to be encrypted "on-demand", I mean, apart from the encryption header data is encrypted "on-the-go" as it is written to the device and ...
- 117
0 votes
1 answer
135 views
Criteria for throttling of Azure disk storage
I need the criteria that will result in throttling of disk storage in Azure. Will high disk IOPS (how high I can see percentage in the metrics) result in throttling? What are the factors that causes ...
- 13
0 votes
1 answer
318 views
How to diagnose slow connection to server that has high upload speed?
I have an Ubuntu server that upon running the speedtest cli showed ~150mbps upload speeds, however when I connect to a librespeed instance running on the server I get only ~4mbps download speeds from ...
- 235
0 votes
1 answer
166 views
More cores = faster response time?
I have a multithreaded application that listens to events and needs to respond to them instantly. With the application running, my server is under 5% CPU load in a 1 core server. I wonder though, ...
- 537
3 votes
2 answers
793 views
ZFS folder very slow after having many files
I am running Ubuntu 24.04 using ZFS for my filesystems. This is on a laptop whose only storage device is a WD Black SN850X NVMe card. The default Ubuntu installation process configured two ZFS pools: ...
- 301
0 votes
1 answer
2k views
Exchange 2019 On-Premise OWA/ECP/Autodiscover Issues
I have lurked here and over at StackExchange (Imagine my surprise to find out they are the same group, just different), and have used questions and answers for years until now when I've hit a ...
- 1
0 votes
1 answer
1k views
Intel turbo boost all core clock speed in practice
My understanding is that Intel's quoted "max turbo boost" speed is only guaranteed when a single core is active, and that when all cores are active, the clock speed could conceivably fall to ...
- 109
2 votes
1 answer
834 views
Intermittent High CPU Spikes and Network Latency on Proxmox VMs Running Kubernetes
Summary: I recently migrated a Kubernetes cluster from bare metal to a Proxmox VM on the same physical host. Since the migration, the VM experiences intermittent high CPU spikes and network latency, ...
- 183
0 votes
0 answers
23 views
Pretesting traffic on a website
I had a php application last year that was working on shared hosting that has 1 GB RAM and 1 CPU core. At the time, there were 200 users using the service and the performance was perfectly fine. ...
1 vote
1 answer
634 views
High IO utilization without high read/write
Our server occasionally becomes laggy when writing files. using iostat I noticed that the disk utilization is abnormally high. However, the read/write per second does not seem too high # example ...
- 13
0 votes
0 answers
67 views
Images from file system are loaded very slowly when on VPS, faster on localhost
My / endpoint returns eleven DTOs, each DTO having a String field imagePath, which looks something like this http://localhost:8082/api/get_image/imagename.png (when on VPS, the imagePath looks like ...
- 165
1 vote
1 answer
524 views
Why Swap usage is high for Influxdb: 100% Disk I/O and Swap usage but only 50% memory?
We have an influxdb VM that is constantly under 100% swap. Even if we restart the VM, the swap usage reaches 100% in about 20 minutes. However, memory usage is only about 50%. (The VM has 32 CPU Cores ...
- 151
0 votes
0 answers
337 views
Random CPU + MEMORY spikes issue on Ubuntu server + Docker + PostgreSQL
I've been encountering an issue that has been bothering me for a while now. I have a Google Cloud VM running an Ubuntu 18.04 server with several containers, including a PostgreSQL server, a Redis ...
0 votes
1 answer
287 views
Could a faulty switch cause upload speeds for some to be slow?
We have recently switched our MSP which has caused a lot of problems with our network, but most of that is fixed and we've got one issue remaining: Half the people in our office have poor upload ...
- 1
1 vote
0 answers
1k views
pmie_daily and pmlogger_daily services both failing to start on rhel server
I've searched and searched, and can't find much on this problem. Both the pmie_daily and pmlogger_daily services are failing to start on out RHEL server. They are part of the RHEL "Performance Co-...
- 151
7 votes
2 answers
2k views
Ansible performance issues on a huge group
So, I have this host group, which consist of 35k VMs. Aaaand I need to run a playbook over it. If it does matter - playbook is just a call for community role for installing node_exporter. But I'm ...
- 98
0 votes
1 answer
528 views
iperf3 download and upload test simultaniously giving abnormal results
I have a server (Lenovo using Ubuntu) and 2 clients (both Mac). The server is connected with a Wi-Fi router over Eth interface. The two clients are connected with Wi-Fi router over WLAN interface. All ...
- 1
2 votes
1 answer
598 views
Postfix: high delay in queue manager
Although this question has been asked before, there was no accepted answer (and the quality of answers was not good). For one site where I send emails to, the "in queue manager" delay metric ...
- 23.9k
0 votes
2 answers
323 views
Debian 12 higher disk activity than CentOS 7
We are in process of replacing our CentOS 7 boxes to run using Debian 12 and have noticed that the overall disk activity on the same server with the same processes running is significantly higher with ...
0 votes
0 answers
42 views
To host a medium to high traffic website would I be better off purchasing 1x32-64 core XEON/EPYIC Server or 20x older i5/i7 desktop quadcore machines?
I'm looking to host a website that could get anywhere between 200-20,000 active users. I'm wondering if I should buy a modern dedicated server with dual socket and a lot of cores (32-120) to run ...
1 vote
2 answers
116 views
How to find the underlying device for a file on top of ext4 on top of LVM on top of multiple underlying storage devices?
I have a system with 3 SSD devices (/dev/sda, /dev/sdb, /dev/sdc) which contain a single LVM logical volume that span over all the devices. I have a single ext4 partition on the logical volume. I ...
- 1,332
-1 votes
1 answer
1k views
Very slow performance with Nginx and Traefik
We have the following infrasctuture to deploy a PHP application: A VM running Nginx as a proxy 3 nodes in a Docker Swarm cluster Traefik containers deployed in this cluster (so we have 3 containers ...
- 109
9 votes
2 answers
2k views
AWS Graviton Processors: which encryption specs are hardware accelerated?
In testing Graviton instances there seems to be significantly more CPU overhead from AES/SHA functions compared to the same on cheaper/smaller Intel/AMD instances. Do these procs supports hardware ...
- 283