TrustRadius
The IBM AC922 and LC922 line of servers are storage-rich servers designed to support high data throughput and scalability needs.https://media.trustradius.com/product-logos/o8/et/WM0B3I7B2EBN.JPEGThe IBM AC922 AI system is training us too!IBM AC922 is currently in use at QCM Technologies for technology exploration, product education, and client engagement. It provides a platform that can be used across our internal team for education and model training. Clients can use it to perform proof of technology exercises as well.,IBM AC922 at QCM is used to train a variety or ML / DL models and provides an exceptional hardware platform for our learning needs. IBM AC922 architecture has shown its ability to shorten the time to value when developing AI models. IBM AC922 has the capability to effectively provide for deployment of an inference engine for vision-related models.,IBM AC922 could benefit from some additional "on system" storage options. It has enough for general use but at larger data scale, external storage is required.,10,Our purchase of IBM AC922 was for educational and demo purposes, so no significant ROI calculations were needed. The system has given us the environment to accomplish our goals. The cost of entry was sufficiently low. That attribute allowed us to get a system for exploration.,The relative speed with which this system executes model training has allowed us to explore multiple model variations and tune parameters in a very time effective manner. In our situation, we don't get the luxury of dedicating time to these tasks and babysitting training runs is complicated for us. IBM AC922 has allowed us to accomplish our goals in a very timely fashion.,In our scenario, we have yet to exploit this capability. We recognize the functionality and we plan to build models to see how this performs.,Best infrastructure for AI workloadWe use IBM AC922 servers as a GPU based infrastructure for a deep learning use case. We leverage it to address use cases across industries and it's leveraged across the USA.,Great on-premise infrastructure for AI. Only GPU infrastructure which uses PCIe Gen 4 and has a Nvlink between GPU and CPU. Comes bundled with WML Base which could be used by data scientists.,Product documentation needs to be kept at pace with changes with the various libraries that need to be installed. No other changes are required in AC922 and we are happy with the infrastructure.,10,Great infrastructure for Enterprise DL use cases. The ROI depends on the use case, but with the top-notch performance it provides, customers will realize the immense value in a period of just 2 years.,We have been using AC922 more specifically for an AI workload and it delivers 20% faster performance as compared to cloud-based GPUs.,As explained earlier IBM AC922 is the best fit for huge data and LM as it supports LMS and it is the best in its class compared to other products we have evaluated.,Enlightening the world on how to do AI in industryIBM AC922 is being used in my organization for machine and deep learning modeling. We find it is best uses are the GPU and NVLink capabilities.,Fast machine learning - It works in minutes instead of hours. Better capabilities with NVLink in comparison to Intel x86. Fabric capabilities are great in a cluster.,More software applications need to be written to gain full use of IBM AC922 and its capabilities. SAS Institute is finally releasing SAS Viya on this platform later in 2019.,9,We have been able to reduce overall computing costs. It has sped up model training times due to it being the fastest architecture currently available.,Yes it has. The most challenging part is finding the business use cases where the use of neural networks can more accurately predict a result. The biggest issue is internal politics. The architecture can do what it needs to do, but the people do not always think outside of the box. Management does not understand it.,Yes, our organization has benefited from this capability.,Amazingly Powerful Architecture for GPU-based ML TasksWe’re currently using the AC922 across the organization in many of our research groups, prototyping and implementing workflows for ML-driven image analysis. We also use the system for genomics and statistical genetics applications on high-throughput sequencing data.,Fast access to in-memory data, speeding up GPU-based computation as compared to commodity hardware. Easy to manage, with a powerful web-based management interface. Integrates remarkably well with other offerings from IBM, including the ESS storage system.,Currently, we’d like to see more support for the Power9 (ppc64le) architecture among popular ML and bioinformatics packages, and lessen the need to compile packages ourselves to run on Power9.,8,We are fairly new to the AC922, but so far we have seen a speed-up on some of our image analysis tasks. Also, the integration with ESS has made it easier to migrate in data.,,AutoCAD mobile (formerly AutoCAD 360), Onshape, Tableau ServerLinux, Red Hat Enterprise Linux, Ubuntu Linux
IBM AC922
12 Ratings
Score 8.4 out of 101
<a href='https://www.trustradius.com/static/about-trustradius-scoring' target='_blank' rel='nofollow noopener noreferrer'>trScore algorithm: Learn more.</a>TRScore

IBM AC922 Reviews

IBM AC922
12 Ratings
<a href='https://www.trustradius.com/static/about-trustradius-scoring' target='_blank' rel='nofollow noopener noreferrer'>trScore algorithm: Learn more.</a>
Score 8.4 out of 101

Do you work for this company?

Overall Rating

Reviewer's Company Size

Last Updated

By Topic

Industry

Department

Experience

Job Type

Role

Reviews (1-4 of 4)

Companies can't remove reviews or game the system. Here's why.
Michael Drapkin profile photo
Score 10 out of 10
Vetted Review
Reseller
Review Source
IBM AC922 is currently in use at QCM Technologies for technology exploration, product education, and client engagement. It provides a platform that can be used across our internal team for education and model training. Clients can use it to perform proof of technology exercises as well.
  • IBM AC922 at QCM is used to train a variety or ML / DL models and provides an exceptional hardware platform for our learning needs.
  • IBM AC922 architecture has shown its ability to shorten the time to value when developing AI models.
  • IBM AC922 has the capability to effectively provide for deployment of an inference engine for vision-related models.
  • IBM AC922 could benefit from some additional "on system" storage options. It has enough for general use but at larger data scale, external storage is required.
IBM AC922 is almost purpose-built for developing AI models. It is extremely capable for this task. It is the core system component in the current fastest "SuperComputer" and a single node for us, so it allows us to get the same experience on a much smaller scale.
Read Michael Drapkin's full review
No photo available
Score 10 out of 10
Vetted Review
Reseller
Review Source
We use IBM AC922 servers as a GPU based infrastructure for a deep learning use case. We leverage it to address use cases across industries and it's leveraged across the USA.
  • Great on-premise infrastructure for AI.
  • Only GPU infrastructure which uses PCIe Gen 4 and has a Nvlink between GPU and CPU.
  • Comes bundled with WML Base which could be used by data scientists.
  • Product documentation needs to be kept at pace with changes with the various libraries that need to be installed.
  • No other changes are required in AC922 and we are happy with the infrastructure.
If there is a huge data repository that is both structured and unstructured, IBM AC922 is the best fit when compared to its competitors. It's like comparing a race between an F1 car and a NASCAR. If the models built are not large (LMS) and DL is not required then IBM AC922 would be overkill.
Read this authenticated review
No photo available
Score 9 out of 10
Vetted Review
Verified User
Review Source
IBM AC922 is being used in my organization for machine and deep learning modeling. We find it is best uses are the GPU and NVLink capabilities.
  • Fast machine learning - It works in minutes instead of hours.
  • Better capabilities with NVLink in comparison to Intel x86.
  • Fabric capabilities are great in a cluster.
  • More software applications need to be written to gain full use of IBM AC922 and its capabilities.
  • SAS Institute is finally releasing SAS Viya on this platform later in 2019.
IBM AC922 is less appropriate for large businesses with hundreds or thousands of business intelligence users. The IBM 980 series seems to make more sense in those applications.
Read this authenticated review
No photo available
Score 8 out of 10
Vetted Review
Verified User
Review Source
We’re currently using the AC922 across the organization in many of our research groups, prototyping and implementing workflows for ML-driven image analysis. We also use the system for genomics and statistical genetics applications on high-throughput sequencing data.
  • Fast access to in-memory data, speeding up GPU-based computation as compared to commodity hardware.
  • Easy to manage, with a powerful web-based management interface.
  • Integrates remarkably well with other offerings from IBM, including the ESS storage system.
  • Currently, we’d like to see more support for the Power9 (ppc64le) architecture among popular ML and bioinformatics packages, and lessen the need to compile packages ourselves to run on Power9.
Operationalized AI pipelines, with high throughput and repeatability, are the ideal scenario for the AC922. Also, GPU-heavy workloads that require large in-memory data sets are what this server does very, very well. Somewhat less appropriate is running libraries that are heavily or exclusively x86-focused.
Read this authenticated review

IBM AC922 Scorecard Summary

About IBM AC922

IBM® Power System™ AC922 provides the data and compute-intensive infrastructure needed to deliver faster time to insights. Power AC922 enables the cutting-edge AI innovation data scientists desire, with the dependability IT requires. This is IT infrastructure redesigned for enterprise AI.
Categories:  Rack Servers

IBM AC922 Features

Has featureFaster I/O - up to 5.6x more I/O bandwidth than x86 servers
Has featurePCIe Gen4 – 2x the bandwidth of PCIe Gen3
Has featurePOWER9 processor—the latest POWER processor, designed for AI
Has featureAdvanced GPUs - up to 6 NVIDIA® Tesla® V100 GPUs with NVLink
Has featureCoherence - share RAM across CPUs & GPUs
Has featureBuilt scalable from one server to supercomputer

IBM AC922 Video

Get into the AI and HPC race with the Power AC922

IBM AC922 Competitors

Dell C4140, HPE Apollo 6500 gen10, NVIDIA DGX-1, NVIDIA DGX-2, Cray CS Storm 500NX

IBM AC922 Support Options

 Free VersionPaid Version
Phone
Live Chat
Email
FAQ/Knowledgebase
Social Media
Video Tutorials / Webinar

IBM AC922 Technical Details

Deployment Types:On-premise
Operating Systems: Linux, Red Hat Enterprise Linux, Ubuntu Linux
Mobile Application:No