This post is an update to the previous post announcing preview availability of the HPCBOX AutoScaler.
Low Priority Instances
As of 2021-08-17, the HPCBOX AutoScaler includes support for low priority instances. Low priority instances are called by different names on different cloud platforms, spot instance on Azure and AWS, preemptible instance on Oracle OCI.
The general idea being, these instances are the same hardware configuration as the standard instances, but, they can be preempted at any time with a short notification by the cloud vendor. Although these instances can be evicted, they do offer a much lower price and are well suited for jobs which don't have a tight deadline.
An important functionality for efficient use of low-priority instances is auto-selection and auto-rescheduling. Auto-Selection meaning, we want the right low-priority hardware to be selected based on the job type, for example, we would like to use specific GPGPU nodes for CUDA jobs, dense CPU nodes with high speed interconnect for CFD jobs, etc. Auto-rescheduling means, we would like the system to be able to automatically reschedule jobs when the cloud vendor is about to evict the instances and for applications which support it, we would also like the jobs to be automatically restarted from the last save point.
We are pleased to announce that the HPCBOX AutoScaler supports both these critical functions out of the box.
Let us consider a use-case to understand how the Intelligent AutoScaler in HPCBOX handles optimization and efficiently handles combining three different classes of hardware on the same cluster.
The following picture represents an HPCBOX Cluster which is a combination of both reserved (resources with a usage commitment, on Azure called Reserved Instances), standard pay-as-you-go resources and low-priority resources. To optimize the budget spend in such a configuration, one would want the reserved instances to be always powered-on to provide a baseline capacity for the cluster and automate the use of pay-as-you-go+low-priority resources to minimize resource wastage. Furthermore, we could assume that the compute workers on this cluster are a combination of different hardware configurations, for example, on Azure, we could assume a combination of HB120rs_V2 and HB120-16rs_v3 (combination of AMD EPYC “Rome” and “Milan” hardware).
Depending on the type of job that comes into the system, AutoScaler either takes no action, powers on standard rate PAYG hardware, or, low-priority PAYG hardware. It also handles auto power-off of the instances once jobs leave the cluster.
When the cloud backbone decides that it needs additional hardware for users who are willing to pay more, our low-priority workers are going to get evicted. However, the AutoScaler makes sure that user jobs get automatically rescheduled without any manual intervention.
Now, comes the best part of the AutoScaler. HPCBOX can automatically attempt restarting of the evicted nodes and when successful, user jobs which got evicted will automatically restart! All this is done with no manual intervention and this, we think, is really how an Intelligent HPC Cloud system should help users optimize their resource usage, deadlines and budgets!
HPCBOX AutoScaler is now available in preview and we would be very pleased to run a demo or perform a POC or pilot with you to optimize your cloud spend on HPC resources while making sure your jobs are always matched to the most suitable hardware. Schedule a meeting here.
Founder and CTO, Drizti Inc
All third-party product and company names are trademarks or registered trademarks of their respective holders. Use of them does not imply any affiliation or endorsement by them.
This is one of the most exciting posts I’ve written in my 15+ years of existence in the High Performance Computing (HPC) industry.
AMD EPYC Zen 3 “Milan” is integrated, tested and available on the HPCBOX platform on Microsoft Azure, on LAUNCH DAY!
This is something which seldom happens in the HPC world, very few users ever get to start production on day one of the release of a new and very impressive processor generation upgrade.
I will not be doing any technical comparison between Zen 3 and Zen 2 and neither will I present any specific application benchmark numbers in this post. I am sure there’ll be many posts by AMD, Microsoft and other ISVs who’ll be publicly sharing information on the performance boost they see for their codes with EPYC Zen 3 “Milan”. This post will be more about HPCBOX and how we could deliver this upgrade experience for our users, on launch day and without having to set our hands on the physical hardware or physically be present in a datacenter(s) (actually multiple regions)! Awesome work and support from the Azure team!
Drizti was a launch partner for the new HBv3 instance size on Microsoft Azure and these instances are powered by the new EPYC 7xx3/Milan CPUs, HDR InfiniBand and sport very impressive dual NVMe drives which give a big performance boost to applications which use local scratch, specifically when they are striped.
Read more about HBv3 here
Unbelievably quick CPU generation upgrade
It all started with Drizti getting access to 1000+ cores of HBv3 instances for functional and compatibility testing, to make sure we are ready for GA availability of HBv3 on launch day of AMD EPYC ”Milan” CPUs. We went through testing, adding necessary support within the HPCBOX platform to make sure we are able to use the cool new features available on the instances, like automate striping of NVMe, testing out MPI compatibility, testing the workflow component of HPCBOX, auto-scaling/shutdown/start etc. and all other functions which are offered by the HPCBOX platform. In addition to this, we also did some performance tests with different ISV and open-source codes, mainly to make sure the HPCBOX workflow engine can correctly handle the new instances and automatically optimize application pipelines to take advantage of the new hardware.
Some of the applications we tested were ANSYS CFX, ANSYS Fluent, OpenFOAM. At a high level, we can share that we are seeing impressive performance benefits of using EPYC “Milan”, particularly for large runs. Also, for applications which are local scratch bound, we expect users to get a good boost due to the possibility of having striped NVMes in HBv3.
We also did a test to see how easy it would be to upgrade HPC for our users and were really impressed that we could just switch our users from HBv2 to HBv3, meaning from “Rome” to “Milan” and to upgraded local scratch in under 30 minutes! This is mainly because of the design of the HPCBOX platform, it is a self-contained platform, and this gives us the ability to fine tune platform capabilities quickly without depending on external services to get upgraded first.
Agile and Impressive!