ClusterVisor · HPC and AI cluster management
Your cluster, under control.
Stop juggling tools. Stop guessing what’s happening inside your cluster. With ClusterVisor, everything you need is right in front of you.
Works on Enterprise Linux clusters from any vendor. Trial licenses available.
Solve problems faster
Optimize resources
Support more users & workloads
Coming soon
ClusterVisor 3.X is coming.
Our biggest update yet: a new web interface, safer cluster builds, and new integrated tools for Slurm, BeeGFS and fabrics.
Inside ClusterVisor
See how it works.
One control point for the whole cluster
The ClusterVisor appliance holds your cluster’s configuration, operating system images and history. A lightweight agent on every node reports health and applies changes, and Slurm, user accounts, logs and hardware management all plug into the same platform, whether we built your cluster or another vendor did.
Meet Provisioner
New in ClusterVisor 2: a complete rewrite that replaces the aging Cloner software, built for stability and manageability. Watch it take a node from power-on to running: network boot, identification, provisioning and ready for work.
See more. Do more. Control everything.
Twelve tools in one platform, from provisioning a new node to tracking every job in the queue.
For any cluster
Runs on the cluster you already have.
Works with your cluster
Built by us, another vendor or your own team, for HPC or AI, ClusterVisor runs on Enterprise Linux 8, 9 and 10 distributions: RHEL, Rocky Linux and AlmaLinux. Slurm is integrated, installed and managed by ClusterVisor.
Secure by design
ClusterVisor supports SELinux in enforcing mode, and its enhanced security option adds full disk encryption and FIPS mode, for research, healthcare and government environments with strict security requirements.
Simple, perpetual licensing
ClusterVisor is licensed per compute node with a perpetual license. Your license gives you access to support and updates. Want to try it first? Trial licenses are available.
Get a quote for your cluster →What’s new
Built for admins. Now for users too.
New
JobVisor
A web interface for researchers to submit jobs, launch interactive desktops and GUI apps, check estimated start times and review job efficiency, without needing to know Slurm. Admins control every option through templates.
Explore JobVisor →New
LogVisor AI
Brings AI to your cluster’s logs, helping you spot, understand and resolve problems faster.
Explore LogVisor AI →Turn cluster data into clarity.
Total visibility
Real-time insight into everything your cluster is doing.
Real-time system and GPU monitoring
Cluster-wide dashboards
Job and queue visibility
Historical analytics
Better performance
Find what’s slowing you down, and fix it.
Identify inefficiencies
Improve scheduling outcomes
Analyze log files
Support system growth
Simpler management
Easier for experienced administrators and newcomers alike.
User-friendly interface
Centralized control
Alerting and notifications
System health insights
Designed for real-world HPC and AI.
ClusterVisor fits into your existing environment and supports the tools and workflows your team already uses, whether we built your cluster or not.
01
Compatible with top schedulers and HPC and AI tools, including Slurm
02
Works across diverse cluster architectures, from CPU-only to GPU-dense
03
Flexible deployment, on a dedicated ClusterVisor appliance or installed on your head node
Why ClusterVisor
Why not stitch it together yourself?
Many teams piece together separate tools for provisioning, monitoring, alerting and scheduling. It works, until an update breaks something, or the person who built it moves on.
Do it yourself
Separate tools for provisioning, monitoring, alerts and scheduling
Integrations you maintain, and repair after every update
Know-how that lives in one admin’s head
Help from forums and mailing lists
ClusterVisor
One interface for provisioning, monitoring, alerts and Slurm
Tested updates that keep everything working together
Consistent, documented management any admin can pick up
Support from the engineers who build it
Case study
Caltech
Professor Marco Bernardi’s team at Caltech is using ClusterVisor on a specialized software development project, where it helped them overcome a RAM bottleneck.
Read the case study →FAQ
ClusterVisor questions, answered.
Have a question that isn’t here? We’re happy to talk it through.
Ask us →Does ClusterVisor work on clusters you didn’t build?
Yes. It runs on clusters built by any vendor or by your own team, on Enterprise Linux 8, 9 and 10 distributions: RHEL, Rocky Linux and AlmaLinux.
Does ClusterVisor work on GPU and AI clusters?
Yes. ClusterVisor provisions, monitors and manages GPU nodes just as it does CPU nodes, including each NVIDIA GPU’s usage, memory and power, and the Slurm it installs and manages schedules jobs across your GPUs.
Can I buy ClusterVisor on its own, without an ACT cluster?
Yes. ClusterVisor is available on its own for the cluster you already have. It’s also included with every cluster we build.
Which job scheduler does it use?
Slurm. It’s integrated with ClusterVisor, which installs and manages it for you.
How is ClusterVisor licensed?
Per compute node, with a perpetual license. Your license gives you access to support and updates.
Can we try ClusterVisor before we buy?
Yes. Trial licenses are available, so you can see ClusterVisor working on your own cluster before you commit.
Ask about a trial →What security features does ClusterVisor support?
ClusterVisor’s enhanced security option adds full disk encryption and FIPS mode, and ClusterVisor supports SELinux in enforcing mode, so your cluster can meet strict research, healthcare and government security requirements.
Can you run ClusterVisor for us?
Yes. With our managed services, our engineers use ClusterVisor to monitor, patch, tune and support your cluster for you.
See managed services →See it on your cluster.
New to ClusterVisor? We’ll walk you through it with one of our HPC/AI engineers.
Request your demo
Already a user? Join COG.
The ClusterVisor Operator Group keeps you up to date on news and events for everyone who runs ClusterVisor.
Join the group
Engineering the future of HPC and AI
info@advancedclustering.com(866) 802-82223148 Roanoke Road, Kansas City, MO 64111