PSSC Labs
Request a Quote

Author: marketing@site-a.com

Understanding HPC and HPC Cluster Management

By marketing@site-a.com

Many of the most significant challenges today in diverse fields such as weather modeling, the federal government, financial services, higher education, life sciences, healthcare, manufacturing, and more are using the power of High Performance Computing (HPC) to create more optimized products and to understand the physical world around us.

Connect With Us

What is the Scope of HPC

HPC has become interwoven into many organizations’ workflows and is only expected to grow over time. With such a wide range of industries using HPC today, designing and implementing an efficient infrastructure requires knowledge of the applications that will be used as well as planning for future growth as well as finding partners that can work with you to determine the ideal solution for your organization. Many organizations may be using HPC in more than one application, which requires the HPC solution to be flexible for different computing, storage, and networking requirements. According to the industry experts at MarketWatch and other sources, the market for HPC servers, storage, middleware, applications, and service is only expected to grow exponentially year over year.

HPC System Components

HPC systems infrastructures contain several subsystems that all must work together efficiently and scale together. The main subsystems of an HPC system are:

  • Compute – This part runs the application, takes the data passed to it, and outputs an answer. Over the past two decades, most algorithms have been parallelized, where parts of a problem are run on separate computers or multiple cores. Each time the results are generated or partial results are generated, they need to be communicated with the other calculations or stored on a storage device. Modern servers contain two to four sockets (chips), each with up to 64 cores. Since each of these cores may need to store recently computed information, the demands on the storage device may increase as core counts increase. Many modern applications now take advantage of accelerators that are tightly connected with the main CPU that can accelerate certain portions of the application. Usually referred to as GPGPUs, these accelerators can speed up specific applications and can place their demands on the storage or networking infrastructure.
  • Storage – During or at the beginning of a long-running simulation, a great deal of data is required to get the simulation going. As the application runs, more data may also be required based on the algorithm. The complexity of storage solutions can quickly grow with the complexity of your application so the PSSC Labs team works with you to make sure the storage is customized to meet your unique needs. The key point is that HPC storage is a key component for the smooth and efficient running of any HPC cluster.
  • Networking – The communication between servers and storage devices should not slow down the performance of the entire system. Each core that is performing the computations may need to communicate with numerous other cores, and request information from other nodes in the process. The network needs to be designed to handle server-to-server communication as well as multiple servers concurrently communicating to the storage system.
  • Application Software – Typically, sophisticated software simulates physical processes and runs across many cores. HPC application simulators are not just complex because of the mathematics behind them, but also because highly tuned libraries are used to manage networking, work distribution, and storage processes. The HPC application software needs to be architected to keep the overall system busy to maximize the investment for an HPC infrastructure.  To learn more about HPC application software click here.
  • Orchestration – Organizing a large cluster can be challenging for network administrators.  Rarely will the entire system of a large supercomputer be dedicated to a single application. Therefore, application software is needed to allow a network engineer to allocate a certain number of servers, GPUs, network bandwidth, and storage capacities. All of these sub-systems, as well as the installation of an Operating System and the associated software on the allocated system need to be handled seamlessly and effortlessly. For applications that require rapid access to data, it is absolutely crucial to correctly set up the software for HPC storage.

What is HPC Cluster Management?

HPC Clusters are composed of components that enable applications to be executed. Therefore, the HPC cluster software will typically run across many nodes and access storage both for reading and writing data. All of this, including the communication between nodes and the storage system, needs seamless communication. According to the European Grid Infrastructure, “High Performance Computing” (HPC) requires distributed computing between many nodes to deliver efficient computational performance, which is typically described as HPC Clusters. Typically, several different types of nodes comprise an HPC Cluster. HPC cluster components will include:

  • Head Node (i.e. the login node) – Often, a head node is nothing more than a simple setup designed to act as a bridge between the cluster and an external network.
  • Compute Nodes – These nodes perform the numerical computations and will probably contain the highest clock rates available with the maximum number of cores at the given clock rate. According to Alex Lesser, a VP at PSSC Labs, “the persistent storage on these nodes may be minimal, while the DRAM memory will be high.”
  • Accelerator Nodes – Some nodes may include one or more accelerators since not all applications can take advantage of these accelerators. In addition, smaller HPC clusters, designed for a specific application may be set up where all the nodes contain an accelerator
  • Storage Nodes (i.e. storage system) – An efficient HPC cluster will need to contain a high performance, parallel file system or PFS. Parallel file systems allow all nodes to communicate with storage drives in parallel. HPC storage enables the compute nodes to operate with minimal wait times.
  • Network Fabric – In HPC clusters, typically, an Infiniband or another high performing Ethernet network switch will be used, due to the requirement of low latency and high bandwidth features.
  • Software – Cluster management software is required to control the underlying infrastructure, as well as the applications to run on the cluster. Managing the multiple I/O streams that are inherent to HPC applications requires the appropriate cluster management software for an optimal HPC cluster.  Parallel transfer of data from the large number of CPUs to the storage system (on the HPC servers or externally) is critical and must not be overlooked.

To learn more about HPC cluster management and HPC servers built for the highest possible performance and reliability including support for Intel Xeon, AMD Epyc, NVIDIA GPU, and Mellanox Infiniband click here https://pssclabs.losangeles.dev.buckupstudio.com/products/ai-hpc-servers/

Keys to Providing the Federal Government with Customized HPC and AI Solutions

By marketing@site-a.com

In order to achieve maximum performance at the lowest possible investment, the PSSC Labs team has been working with federal government agencies to develop a deeper understanding of what is important as well as their technical specifications.  This consultative approach is the key to exceeding the needs of the federal government.

Connect With Us

“Over our 20+ years working with federal agencies, we understand that a one-size-fits-all approach simply does not work. Each organization and group has a specific mission requiring specific computing and storage needs.  We work closely with the end users to ensure they receive the right platform,” said Alex Lesser, VP for PSSC Labs.

PSSC Labs is trusted as a proven HPC and AI/DL technology partner – with the Department of Defense, NASA, the National Oceanic and Atmospheric Administration (NOAA), and the National Institutes of Health (NIH). PSSC Labs is uniquely qualified to help federal agencies embrace HPC / AI technologies. We have deep expertise with federal agencies in these HPC and AI systems, which are rapidly becoming the solutions for agency computing needs.

PSSC Labs HPC / AI solutions are TAA compliant and innovative, while being cost-effective and delivered on time.

Contract Vehicles

PSSC Labs offers HPC and AI products through a variety of government contract vehicles. These widely accepted contracting vehicles expedite the purchasing process while satisfying government purchasing regulations. In addition, PSSC Labs is classified as a small, woman-owned business.

Federal Government Contract Vehicles

GSA Schedule # GS-35F-0598R
US Army Chess
CMAS (California Multiple Award Schedule)
NASA SEWP
NIH NITAAC

Major Benefits of Using Object Storage

By marketing@site-a.com

It is difficult to pinpoint an exact definition of object storage since the concept has different meanings within different organizations. In its most basic explanation, object storage is a system that places your stored data on a flat plane called a “storage pool.” Unlike other storage systems, objects are not kept in a hierarchy and cannot be put inside one another. This is beneficial for a number of reasons. Connect With Us “Object storage offers a simple way to access files any time, anywhere from any device. In today’s world of remote work, object storage can be the key to increasing productivity for organizations of all sizes. In addition, object storage can significantly reduce the overall cost of storage by allowing organizations to break free from proprietary and expensive tier 1 storage platforms,” says Alex Lesser, VP of PSSC Labs. Object storage offers a simple way to access files any time, anywhere from any device. In today’s world of remote work, object storage can be the key to increasing productivity for organizations of all sizes. In addition, object storage can significantly reduce the overall cost of storage by allowing organizations to break free from proprietary and expensive tier 1 storage platforms.

4 Benefits of Object Storage

1. Scalability Object storage is known for its compatibility with cloud computing, and that’s because of its unlimited scalability. Thanks to its flat structure, object storage doesn’t have the same limitations as file or block storage. Hierarchical systems often experience complications when it comes to scaling out. Those difficulties are avoided in an object storage environment because the system scales out by adding nodes. Object storage can easily scale data to petabytes without restrictions. Though it has these advantages in scalability, it doesn’t mean object storage systems can’t also benefit static data. In fact, object-based storage is optimal for static data. For this reason, object storage has a sort of niche applications in the market, best serving static yet scalable archival storage, according to ComputerWeekly. For example, object storage could be ideal for digital archives. 2. Faster Data Retrieval and Better Recovery Each object in the storage environment has its own identifying details, comprised of metadata and ID number, which the OS reads to retrieve data. Without the need to sift through file structures, retrieval is much faster. Thanks to the metadata and ID numbers, users don’t need to know an object’s exact location to retrieve it. TechTarget compares object storage to valet parking: you don’t know where your car is parked, or if and how many times it moves throughout the evening, but by having a receipt, you’ll be able to retrieve your car at the end of the night. Additionally, having unrestricted metadata allows storage administrators to implement their own policies for data preservation, retention, and deletion. This, along with the way storage nodes are distributed across the structure, makes it easier to reinforce data and create better disaster recovery strategies. 3. Fewer Limitations Compared to the traditional file or block-based systems, object storage is far less limited because it’s not organized in a hierarchy. Object storage provides the kind of access that other storage systems can’t allow. TechTarget also points out that the metadata in file systems is limited to file attributes, whereas, in object storage, it can be customized by any number of data attributes. It’s only possible to achieve that in file systems through a separate application. In addition to these advantages, as well as its unrestricted scalability, this all contributes to a limitless storage system that files or blocks can’t provide. 4. Cost-effectiveness For organizations that need to store large amounts of data, an object-based system could be the most cost-effective. Because it scales out much easier than other storage environments, it’s less costly to store all your data. Plus, if users have a private cloud space, costs can be even lower. When compared to other systems that are considered inexpensive for these volumes of data, it’s a much more durable alternative. Learn More To learn more about object storage and implications for your organization contact us at 4sales@site-a.com or 949-380-7288.  For over 25 years PSSC Labs has been providing custom, on-premise high performance computing (HPC) systems designed to meet our clients’ unique enterprise computing challenges.  So whether you need help with an HPC AI or Deep Learning solution or other Big Data application, PSSC Labs is here for you and your organization.
To download the infographic as a PDF, please click here.
Benefits of Using Object Storage

PSSC Labs Named Ansys Preferred Solution Partner

By marketing@site-a.com

LAKE FOREST, CALIF., July 23, 2020 — PSSC Labs, a developer of high performance computing (HPC) and big data computing solutions for design & engineering, today announced that Ansys, the global leader in engineering simulation software, has named PSSC Labs an Ansys Preferred Solution Partner. Through this partnership, PSSC Labs will provide custom configured computing solutions along with best-in-class customer service to Ansys end users.

“As hardware increases in complexity and diversity, it can be difficult for clients to select the most appropriate hardware for their engineering simulation workloads,” said Wim Slagter, director of HPC and cloud alliances at Ansys. “PSSC Labs has addressed this challenge by its ability to build custom solutions for even the most challenging Ansys workloads, each built for reliability, security and performance.”

PSSC Labs is already working with several leading government and commercial organizations to deliver HPC solutions specifically for Ansys applications.

“This partnership validates what we have been doing for many years already, delivering turn-key custom platforms for Ansys workloads,” begins Alex Lesser PSSC Labs Vice President. “PSSC Labs is really unique in our understanding of complex design and engineering workflows.  Rather than just offer a canned server or cluster, we carefully listen to the needs of each individual and work with them to deliver the right solution on-time and on-budget.”

PSSC Labs can pre-install all necessary applications through the Application Integration Service (AIS). This is a big win for PSSC Labs and Ansys clients as they can begin immediately running their workloads and simulations on a validated platform by the manufacturer.

As an Ansys Preferred Solution Partner, PSSC Labs will develop and invest in new technologies that ensure the highest performing computing solutions specifically for complex solvers and other Ansys applications. PSSC Labs’ PowerWulf ZXR1+ HPC clusters and PowerServe Uniti servers are already proven to be compatible with Ansys applications. The goal now is to continue pushing the limits of computational performance while maintaining the lowest total cost of ownership. All PSSC Labs platforms can be customized to meet specific needs, budgets and timelines. PSSC Labs engineers listen carefully to specific needs and can craft highly customized solutions to meet them.

As an Ansys Preferred Solution Partner, PSSC Labs can leverage their success on several projects within the military and government spaces for clients such as the U.S. Army, U.S. Air Force, U.S. Navy, NASA, Fortune 500 companies, and small and medium sized businesses. 

About PSSC Labs

For technology powered visionaries with a passion for challenging the status quo, PSSC Labs is the answer for hand-crafted HPC and Big Data computing solutions that deliver relentless performance with the absolute lowest total cost of ownership. We are true innovators offering high performance computing solutions to solve the world’s most demanding problems. For 25+ years, organizations of all sizes and from a variety of sectors rely on PSSC Labs’ computing systems. We are proud to support many departments within the United States government, Fortune 500 companies, as well as small and medium-sized businesses.

All products are designed and built at the company’s headquarters in Lake Forest, California.

Machine Learning vs. Deep Learning: What’s the Difference and Why Does it Matter?

By marketing@site-a.com

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 Steps Up to Bat Following AMD EPYC Milan Processor Release: Enter Ice Lake

By marketing@site-a.com

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.

CPU Name Cores/Threads Base Clock Boost Clock (1-Core) Boost Clock (All-Core) L3 Cache L2 Cache TDP
Xeon Platinum 8380 40/80 2.30 GHz 3.40 GHz 3.00 GHz 60 MB 50.00 MB 270W
Xeon Platinum 8368Q 38/76 2.60 GHz 3.70 GHz 3.30 GHz 57 MB 47.50 MB 270W
Xeon Platinum 8368 38/76 2.40 GHz 3.40 GHz 3.20 GHz 57 MB 47.50 MB 270W
Xeon Platinum 8362 32/64 2.80 GHz 3.60 GHz 3.50 GHz 48 MB 40.00 MB 265W
Xeon Platinum 8360Y 36/72 2.40 GHz 3.50 GHz 3.10 GHz 54 MB 45.00 MB 250W
Xeon Platinum 8352S 2.20 GHz 3.40 GHz 2.80 GHz 48 MB 40.00 MB 205W
Xeon Platinum 8352Y 32/64 2.20 GHz 3.40 GHz 2.80 GHz 48 MB 40.00 MB 205W
Xeon Platinum 8358 32/64 2.60 GHz 3.40 GHz 3.30 GHz 48 MB 40.00 MB 250W
Xeon Platinum 8358P 32/64 2.60 GHz 3.40 GHz 3.20 GHz 48 MB 40.00 MB 240W
Xeon Platinum 8352V 36/72 2.10 GHz 3.50 GHz 2.50 GHz 54 MB 45.00 MB 195W
Xeon Platinum 8351N 36/72 2.40 GHz 3.50 GHz 3.10 GHz 54 MB 45.00 MB 225W

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.