Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register

Quick Links


Shells Virtual Desktop
BMail.ag - Secure Email Service
Server.net
CPLicense.net
VPS Server
Buy VPN
Vultr
VMs for AI
HostDare
ReliableSite White-Label Dedicated Hosting for Resellers
InterServer VPS
BMail.ag - Secure Email Service
Best VPN
High-Performance Bare Metal Server Solutions
Karvl.com
Server Mania Cloud Hosting
DataWagon Hosting
AlphaVPS Hosting
Evoxt.com
Clouvider
VPS Hosting with NVMe
Residential IPs in the US & 4G Mobile Proxies in EU & US with Unlimited Bandwidth
ReliableSite White-Label Dedicated Hosting for Resellers
Rabisu - Hosting Solutions
Shells Virtual Desktop

Categories

In this Discussion

Home › Offers
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

GreenCloud | 10th Birthday Sale | 1010 plans | Giveaways

1…8384868889…113

Comments

  • OKHaoOKHao Member

    17 Available 2222 SJC

  • bdlbdl Member

    @I_Won_All said:
    What's the discount for your birthday?

    minus one year before your deathday

  • nyamenknyamenk Member

    MOARRRRRRRRRR

  • rincewindrincewind Member

    @ehab said: I am late to the game but better i start from the experienced before its too late. Thanks in advance.

    OpenAI is great but GPT-4 is expensive, and the free version often hallucinated (shitty/misleading output). There are also open source versions like StarCoder and CodeLLaMA (https://ai.meta.com/blog/code-llama-large-language-model-coding/) that you can run on a GPU.

    But for practical queries, I have switched to using https://perplexity.ai/ as my favourite search engine. It has a CoPilot which is still free if you create an account, and the Pro version connects to GPT-4.

  • codelockcodelock Member

    Only 14 more pages we can do this

  • kaz050457kaz050457 Member

    Happy 10th Birthday!

  • GanonkGanonk Member

    go to 💯 pages

  • phiexzphiexz Member

    Happy 10th Birthday!

  • loayloay Member

    @TODO said: Do you mind enlightening me about that area, I am a hobbyist so I never really touched S3 as I see it is much more expensive than a storage VPS or a pcloud LTD in the long term.

    I use idrive e2 with 1TB storage for 20$ per year then just mount it in the Jellyfin server with local caching enabled, I posted about this setup here:

    https://lowendtalk.com/discussion/comment/3725563#Comment_3725563

  • OKHaoOKHao Member

    1 Available EPYCSG-1
    8 Available EPYCJP-1

  • loayloay Member

    @rincewind said: Using cgroups is better than cpulimit.

    The idea is that I don't to check for this manually or run specific commands. I want to set it system wide so cpulimit is more useful in this case, I think proxmox and docker which was mentioned earlier but I could not find that comment from the spam here lol

    https://docs.docker.com/config/containers/resource_constraints/#configure-the-default-cfs-scheduler

  • bapbap Member

    Happy 10th Birthday!

  • rincewindrincewind Member

    @loay said: cpulimit is more useful in this case,

    Docker internally uses cgroups, so its the same thing without the overhead of a container. The script you generated doesn't actually work - it runs cpulimit -l 30 for each process. So if you had three ffmpeg's running it would restrict each to 30% and total utilization would jump to 90%

    Doing on a systemwide basis is difficult.

    Thanked by 2loay dev077
  • loayloay Member

    @rincewind said: Doing on a systemwide basis is difficult.

    :(

  • JackLiuJackLiu Member

    Happy 10th Birthday! I have bought your service. Perhap i can do it.

  • rincewindrincewind Member

    @loay said: Doing on a systemwide basis is difficult.

    Maybe there is a way: You could use cgclassify to move all cpu-intensive tasks to the same cgroup. So adjust your script and change the cpulimit line to cgclassify -g cpu:cpulimited $pid

    Thanked by 1loay
  • pkrpkr Member
    edited October 2023

    I think those who could not get 1010 need to wait for BF deals.

  • loayloay Member

    @rincewind said:

    @loay said: Doing on a systemwide basis is difficult.

    Maybe there is a way: You could use cgclassify to move all cpu-intensive tasks to the same cgroup. So adjust your script and change the cpulimit line to cgclassify -g cpu:cpulimited $pid

    I actually don't have any experience with this but here is chatgpt comment and modified scripts. I still don't know if it will work.

    You're absolutely right, and I appreciate the suggestion. Using cgclassify to move CPU-intensive tasks to the same cgroup can be an effective way to control CPU usage collectively for those tasks. Here's how you can adapt the script:

    1. First, create a cgroup for your CPU-limited tasks if you haven't already:

      sudo cgcreate -g cpu:cpulimited

    2. Modify the script to use cgclassify to move processes to the cpulimited cgroup and enforce CPU limits:

      #!/bin/bash

# Define the burst duration in seconds (10 minutes = 600 seconds)
BURST_DURATION=600
# Define the maximum burst time allowed in seconds (24 hours)
MAX_BURST_TIME=86400
# Initialize variables to track burst and last burst time
BURST_START=0
LAST_BURST=0

while true; do
 # Get the current time in seconds since midnight
 CURRENT_TIME=$(date -d "now" +%s)

 # Check if a new burst period can begin
 if ((CURRENT_TIME - LAST_BURST >= MAX_BURST_TIME)); then
   BURST_START=0
 fi

 # Check if a burst period is allowed
 if ((BURST_START == 0 || (CURRENT_TIME - BURST_START) <= BURST_DURATION)); then
   # During the burst period, move processes to the cpulimited cgroup
   cgclassify -g cpu:cpulimited -r "$$"  # Move the current script to the cpulimited cgroup
   CPU_LIMIT=100
   BURST_START=${BURST_START:-$CURRENT_TIME}
 else
   # Outside the burst period, limit the CPU usage to 30%
   CPU_LIMIT=30
   cgclassify -g cpu:cpulimited -r ""  # Remove processes from the cpulimited cgroup
 fi

 # Find all running processes and apply CPU limits using cpulimit
 for PID in $(pgrep -d, -x .); do
   cpulimit -l $CPU_LIMIT -p $PID
 done

 # Update the last burst time if a burst occurred
 if [ "$CPU_LIMIT" -eq 100 ]; then
   LAST_BURST=$CURRENT_TIME
 fi

 # Sleep for a while before checking again (adjust as needed)
 sleep 5
done

    This updated script moves processes to the cpulimited cgroup during the burst period and removes them from the cgroup when outside the burst period. This way, it collectively manages CPU usage for the specified tasks within the cgroup while maintaining the 30% limit for the rest of the system.

  • nyamenknyamenk Member

    MOARRRRRRRRRR

  • kaz050457kaz050457 Member

    Happy 10th Birthday!

  • QuartxQuartx Member

    Almost forgot about this , lucky i checked on the finalish day

  • kainjinezkainjinez Member

    MOARRRRRRRRRR

  • codelockcodelock Member

    Why is this thread almost dead now

  • totototototo Member 
    1 Available - BudgetKVMMO-1
1 Available - EPYCSG-1
2 Available - BudgetKVMIL-2
2 Available - BudgetKVMMO-2
2 Available - BudgetKVMNL-2
2 Available - BudgetKVMNY-1
2 Available - BudgetKVMSJC-3
2 Available - EPYCDE-2
3 Available - BudgetKVMIL-3
3 Available - BudgetKVMJP-2
3 Available - BudgetKVMSJC-2
3 Available - EPYCDE-3
3 Available - EPYCTX-1
4 Available - BudgetKVMJAX-1
4 Available - BudgetKVMMO-3
4 Available - EPYCSJC-2
5 Available - BudgetKVMAZ-3
5 Available - BudgetKVMNL-3
5 Available - EPYCJP-3
5 Available - EPYCUK-3
6 Available - EPYCDE-1
6 Available - EPYCSG-2
6 Available - EPYCSG-3
7 Available - BudgetKVMHK-2
7 Available - BudgetKVMNY-2
7 Available - BudgetKVMSG-2
7 Available - BudgetKVMTX-3
7 Available - EPYCSJC-3
8 Available - BudgetKVMHK-3
8 Available - BudgetKVMTX-1
8 Available - BudgetKVMUK-3
8 Available - EPYCJP-1
8 Available - EPYCJP-2
8 Available - EPYCSJC-1
10 Available - BudgetKVMAZ-2
11 Available - BudgetKVMTX-2
12 Available - BudgetKVMSG-3
14 Available - 2222 SJC
15 Available - EPYCUK-2
16 Available - BudgetKVMJAX-2
18 Available - BudgetKVMJP-3
27 Available - BudgetKVMJAX-3
28 Available - EPYCUK-1
  • plumbergplumberg Veteran, Megathread Squad

    @tototo said:

    1 Available - BudgetKVMMO-1
1 Available - EPYCSG-1
2 Available - BudgetKVMIL-2
2 Available - BudgetKVMMO-2
2 Available - BudgetKVMNL-2
2 Available - BudgetKVMNY-1
2 Available - BudgetKVMSJC-3
2 Available - EPYCDE-2
3 Available - BudgetKVMIL-3
3 Available - BudgetKVMJP-2
3 Available - BudgetKVMSJC-2
3 Available - EPYCDE-3
3 Available - EPYCTX-1
4 Available - BudgetKVMJAX-1
4 Available - BudgetKVMMO-3
4 Available - EPYCSJC-2
5 Available - BudgetKVMAZ-3
5 Available - BudgetKVMNL-3
5 Available - EPYCJP-3
5 Available - EPYCUK-3
6 Available - EPYCDE-1
6 Available - EPYCSG-2
6 Available - EPYCSG-3
7 Available - BudgetKVMHK-2
7 Available - BudgetKVMNY-2
7 Available - BudgetKVMSG-2
7 Available - BudgetKVMTX-3
7 Available - EPYCSJC-3
8 Available - BudgetKVMHK-3
8 Available - BudgetKVMTX-1
8 Available - BudgetKVMUK-3
8 Available - EPYCJP-1
8 Available - EPYCJP-2
8 Available - EPYCSJC-1
10 Available - BudgetKVMAZ-2
11 Available - BudgetKVMTX-2
12 Available - BudgetKVMSG-3
14 Available - 2222 SJC
15 Available - EPYCUK-2
16 Available - BudgetKVMJAX-2
18 Available - BudgetKVMJP-3
27 Available - BudgetKVMJAX-3
28 Available - EPYCUK-1

    Wow. How did you scrape this?

  • rincewindrincewind Member

    @loay said: I still don't know if it will work.

    The burstable script generated is awful. The first script was somewhat better, but even that has mistakes : Getting pids+usage should be top -bn1 | awk '{print $1, $9}' | tail -n +8.

    Here's some documentation on cgroups-v2 to get you started : https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/8/html/managing_monitoring_and_updating_the_kernel/using-cgroups-v2-to-control-distribution-of-cpu-time-for-applications_managing-monitoring-and-updating-the-kernel

    Thanked by 1loay
  • zhizhi Member

    Happy 10th Birthday!

  • loayloay Member

    @rincewind said: The burstable script generated is awful. The first script was somewhat better, but even that has mistakes : Getting pids+usage should be top -bn1 | awk '{print $1, $9}' | tail -n +8.

    Okay, thank you! I think I should not depend on chatgpt for this.

    Thanked by 1rincewind
  • OKHaoOKHao Member

    8 Available EPYCSJC-1

  • codelockcodelock Member

    @Ganonk said:
    go to 💯 pages

    Let's gooooooooooooooooooo

Sign In or Register to comment.