At its simplest, machine learning (ML) and deep learning (DP) are both facets of artificial intelligence (AI). To those not in the industry, the terms seem interchangeable, but that’s not quite the case. Deep learning is actually a more advanced, more extensive version of machine learning. To understand further, let’s break down machine learning by itself.

Machine Learning

Machine learning is an application of artificial intelligence that provides systems with the ability to automatically learn and improve from experience and testing, without being explicitly programmed to do so. With ML, computer programs can access data and learn from it themselves. Machine learning involves a lot of complex math and coding, so when we say something is capable of machine learning, we mean that it’s something that can perform a function with provided data and progressively get better over time.

Deep Learning

Deep learning is not wildly different, though its capabilities are. DL imitates the way that the human mind processes information and converts it to knowledge. It’s an important element of data science, which includes statistics and predictive modeling. So how exactly does deep learning work? With the help of a layered structure of algorithms called an artificial neural network. This network is considered the human “brain”. As stated in an article from Zendesk, “It’s a tricky prospect to ensure that a deep learning model doesn’t draw incorrect conclusions—like other examples of AI, it requires lots of training to get the learning processes correct. But when it works as it’s intended to, functional deep learning is often received as a scientific marvel that many consider being the backbone of true artificial intelligence” (Grossfeld, 2020).

The key difference between the two is the differences in performance as the scale of data changes. When a certain data set is small, machine learning is the best fit. But as the data size increases, deep learning becomes necessary because the DL algorithms perform better. This is because a large amount of data is needed for DL to understand and learn from it perfectly.

PSSC Labs has been an expert in the machine learning and deep learning disciplines of computing for 25+ years, providing custom, unique high performance computing solutions to organizations in various industries. To learn how our solutions can help you achieve your business objectives, visit our HPC solutions page and request a quote for your custom system.

Intel has recently released its new Ice Lake Processor Family, the third-generation Intel Xeon Scalable processor that could be just the right HPC, AI or cloud solution for your organization. With up to 40 “Sunny Cove” cores per processor, built-in acceleration and new instructions, the Ice Lake CPUs will offer significant performance boosts for various AI, HPC, and cloud applications/workloads, according to Intel.

Increasing from a 28 core count in the previous generation of the Cascade Lake processor to 40, Ice Lake will provide you with eight channels of DDR4-3200 memory per socket and up to 64 lanes of PCIe Gen4 per socket, rather than the six channels of DDR4-2933 and up to 48 lanes of PCI Gen3 per socket for the previous generation. These changes among others result in a 46% performance improvement for datacenter workloads and a 53% higher average HPC performance, generation-over-generation, according to Intel. “Having eight memory channels is key for memory bound workloads, and with the 40 cores along with AVX-512, the CPU shows great performance for a lot of workloads that are more compute bound,” said Trish Damkroger, Intel’s VP and General Manager of HPC.

What Intel Ice Lake New Processors Mean for Industry Applications

Weather Modeling Applications
These new Intel Ice Lake processors create a distinctive advantage in weather modeling especially for meteorologists concerned about upcoming weather patterns. HPC solutions equipped with these CPUs are going to combine higher core counts and higher speed processors to allow for more and more accurate weather models to be run in a given time period. The ability to run more weather models gives meteorologists and weather researchers the ability to study weather modeling factors that play significant roles in natural disasters, like hurricanes, tropical storms, and challenging wind factors and their impact on wildfires. With the information provided from these models, researchers can work quickly with utility companies, when time is of the essence, to help reduce the risk of natural weather disasters by recommending how and when to shut down operations. Many of our current clients are utilizing the Intel-based PowerWulf ZXR1+ HPC Cluster, which is profoundly advancing the work of meteorologists.  As weather projections for the 2021 hurricane and tropical storm season continue to need HPC weather modeling solutions like these PSSC Labs has the solutions that can help you be prepared for hurricanes or wildfires.  Click here to connect with a weather modeling expert.

Life Sciences and Bioinformatics
Many of our clients operate in bioinformatics and genomics, among many other types of life science research where speed of processing is of the utmost importance. With the new Intel Xeon Scalable Ice Lake Processors, life science researchers can have more processing cores at a very competitive cost per core. The need for speedy computations in the life science space has never been more important, especially with the amount of focus on virology research due to COVID-19 since 2020. We needed bioinformation information and treatment insights, and we needed it fast. Thanks to high performance computing and the right hardware, we’ve reached a point where you have more solutions than ever before. The ability of researchers to deliver solutions due to enough processing cores to handle an ever-increasing amount of data means faster time-to-answers, benefitting us all going forward.  Speak with a Life Sciences and Bioinformatic Professional today.

Design & Engineering
Engineers need the best of the best to perform their work, and the ability to throw significantly more and significantly faster processor cores at their design and engineering models and simulations is ultimately what allows for finer mesh models. With even faster memory access speeds than earlier generations, the Intel Ice Lake processors allow engineers to have much faster access to important data, which is critical when dealing with low latency applications from companies like Metacomp Technologies, Star CCM and Ansys.

Take a look at the PSSC Labs family of products and use our product configurator tool on any of our product pages to see your specific configuration options. The configurator tool can also be used to request a quote, which we’ll return in 24 hours or less.  You can also connect with one of our experts today directly by clicking here.

PSSC Labs builds technology for visionaries seeking handcrafted computing solutions that will take their work to the next level. With the lowest total cost of ownership (TCO) and no need to sacrifice on performance, reliability, or security, the PSSC Labs team and product suite are true innovators offering high performance computing solutions to solve the world’s most demanding problems. Our 25+ years of experience is highlighted in our work with organizations of all sizes and a variety of industries, including many departments within the United States government, Fortune 500 companies, as well as small and medium-sized businesses.

All products are designed and built in the U.S. at the PSSC Labs headquarters in Lake Forest, California.

News of Jeff Bezos’ flight to space hit the airwaves this month with fervor. Self-proclaimed space nerds, tech nerds, and the like swarmed the internet to watch billionaire Amazon tycoon and his crew take flight to the edge of space. He’s not the first and likely won’t be the last.

But before anyone can actually launch into the great beyond, a risk and safety analysis must be performed. This range safety calculates hazards and risks due to spacecraft malfunctions.  In simple terms, what happens to the civilian population if the space craft accidentally explodes; whose house is destroyed. 

To conduct this analysis there are thousands of data points to consider and equal or greater number of potential simulations that must be run.  This requires dedicated, secure, reliable, and consistent high performance computing power.   PSSC Labs on premise PowerWulf ZXR1+ HPC Clusters are employed by the Federal Aviation Authority FAA, United States Navy, United State Air Force, United States Space Force and others to perform this vital step in the mission launch protocol.  In fact no mission to space actually receives the green light for launch until the PowerWulf ZXR1+ HPC Clusters have completed their range safety analysis. 

For over 25 years, PSSC Labs has been providing custom, on-premise high performance computing systems designed to meet our clients’ unique enterprise computing challenges, no matter what they may be. We are experts in deep learning solutions, HPC AI, and big data applications. PSSC Labs is committed to delivering relentless performance with the absolute lowest total cost of ownership all backed with industry-leading US-based support. We’re confident we’ll have the perfect custom computing solution for you. 

For more information on our PowerWulf ZXR1+ HPC Cluster, visit our website and take a look at the cluster tech specs.

AMD recently announced the release of the third iteration of AMD EPYC™   CPUs, touted as “the world’s highest performing server processor” by AMD. Milan hit the market as a replacement for the Rome processor family on March 15th, 2021. Achieving top marks in performance and security, Milan is changing the game for those looking for high-powered processors for their HPC and AI servers.

Notable Characteristics of AMD EPYC Milan CPUs:

• Scaling from 8 to 128 cores
• Up to 16 double-precision FLOPS per cycle per core
• All new “Zen3” cores delivering up to 19% IPC uplift
• Up to 32MB of L3 cache per core, up to 256MG L3 cache per CPU
• New AMD Infinity Guard powering confidential computing
• 4-6-8 memory channel interleaving offering more configuration flexibility
• Support for PCI-Express generation 4.0 (2x the throughput of gen 3.0)
• 128 lanes of PCI-Express 4.0 per CPU socket

The latest generation of AMD EPYC CPUs offer excellent performance, like many of AMD CPU generations in the past. Most models in Milan offer at least 1 TFLOPS, with the 64-core CPUs providing 2+ TFLOPS. Lower-performing models might provide fewer CPU cores and L3 cache memory, while models on the higher end of performance offer higher core counts.

PSSC Labs is also offering free AMD EPYC benchmarking system analysis to further support your decision making process. If you are interested please complete the form below and one of our engineers will be in touch with you.

AMD EPYC™ Milan Processor Tech Specs

What AMD 3rd Gen CPUs Mean for Industry Applications

Weather Modeling
These new AMD EPYC processors offer significant improvements for meteorologists. The key is this solution provides weather modeling systems with higher core counts combined with higher speed processors that allow more models to be run in a given period of time, which is critical for the accurate and timely weather modeling.  The ability to answer questions surrounding wind speed and direction has the potential to impact designated fire risk zones or how organizations should reposition their assets to avoid flooding disasters. The increased processing capabilities of these third series of AMD EPYC series processors have the power to help organizations save lives and protect property. Our current projects are supporting meteorologists leveraging the AMD-based PowerWulf ZXR1+ HPC Cluster, which will have a profound impact on individuals and companies alike.

Life Sciences
Life Science professionals working with bioinformatics or genomics, or any other type of life science research, are always concerned on how many cores they can get in their HPC system. With the new AMD EPYC processors, life science researchers are now able to pack more processing cores into their high performance computing environments at a very competitive cost per core. With the amount of focus being put on virology research due to COVID-19, these third generation AMD EPYC processors stand to play as significant of a role as the researchers themselves. Being able to deliver solutions with enough processing cores to handle an ever-increasing amount of data means faster time-to-answers, benefitting us all going forward.

Design & Engineering
The ability of engineers to throw significantly more and significantly faster processor cores at their design and engineering models and simulations ultimately is what allows for finer mesh models and more models overall. The new AMD EPYC Milan CPUs offer two additional performance enhancements that will significantly improve design and engineering workloads — faster memory access speeds with more memory channels and PCI Express 4.0 buses. Both of these features offer faster access of data to and from processor cores, which is critical when dealing with low latency applications from companies like Metacomp Technologies, Star CCM and Ansys.

Take a look at the PSSC Labs family of  products and use our product configurator tool on any of our product pages to see your specific configuration options. The configurator tool can also be used to request a quote, which we’ll return in 24 hours or less.

PSSC Labs builds technology for visionaries seeking handcrafted computing solutions that will take their work to the next level. With the lowest total cost of ownership (TCO) and no need to sacrifice on performance, reliability or security, the PSSC Labs team and product suite are  true innovators offering high performance computing solutions to solve the world’s most demanding problems. Our 25+ years of experience is highlighted in our work with organizations of all sizes and a variety of industries, including many departments within the United States government, Fortune 500 companies, as well as small and medium-sized businesses.

All products are designed and built in the U.S.  at the PSSC Labs  headquarters in Lake Forest, California.