Servers & Systems: The Right Compute

Optimize memory-centric workloads with enhanced server architecture

The HPE ProLiant DL385 Gen10 server can remove the bottlenecks of memory and I/O with the help of the AMD EPYC processor. Read on to learn how your enterprise can benefit even more.

Blog_EnchancedArchitecture.jpgTo digitally transform, your organization must be able to tackle an increasingly diverse—and incredibly large—amount of data. In other words, your enterprise may have distributed data processing technologies that allow for the creation of large, decentralized data lakes, but can you navigate those lakes smoothly and easily?

If not, you're impeding your digital transformation, and you may want to move away from hard disk-based processing and, instead, rely on memory-centric workloads and computation—an emerging trend among businesses today. However, keep in mind that this shift to memory-centric workloads and computing power comes with a need for increased memory capacity and bandwidth to feed data to processor cores. You'll need to optimize workloads and costs to take full advantage of your computing power and minimize wasted computational resources.

If your organization is looking to pursue this type of initiative, it's imperative to drive for the best value in investment. From superior memory bandwidth and capacity to per core density and I/O, the HPE ProLiant DL385 Gen10 server is proving itself worthy, as it can provide your enterprise with the performance and raw computational power it needs.

Uncover trends in your enterprise's data needs

In choosing a server platform, the first critical step is to understand what your enterprise needs. As the architecture of storage solutions has evolved to meet continuing business challenges, so too must the processing solutions that run on top. Rising to this need are large distributed platforms, which can store a huge variety of enterprise data sets, facilitating both structured and unstructured data.

Memory-centric workloads represent the future of data processing. With raw throughput and order processing of this magnitude, it's faster than traditional I/O and hard disk solutions. Despite this advance, these workloads can be constrained by running on servers that aren't optimized for these uses. Your choice of server hardware and software lays the foundation for modernizing data infrastructure and accelerating business insights.

Discover the right memory architecture to enhance your memory-centric workloads

The HPE ProLiant DL385 Gen10 server represents a platform solution for modern workloads, as a variety of the features present are highly optimized for these specific needs. One of the key aspects setting the HPE ProLiant DL385 Gen10 apart is how it removes the bottlenecks of memory and I/O with the AMD EPYC processor.

The EPYC processor strikes the perfect balance of memory, threads, cores, and raw I/O, specifically built around delivering the most optimized high performance computing power for memory-intensive compute. Implementing a server solution that isn't optimized for these needs can severely destabilize your ability to scale workloads and take advantage of the latest applications designed for memory-centric workloads.

In addition, the design and specifics of the DL385 architecture make it particularly well suited for applications of in-memory databases due to the technical specifications of its memory and processing technology. Regarding RAM, advanced features—such as 4TBs of memory—allow the HPE ProLiant DL385 Gen10 to shine when it comes to providing a highly optimized environment for in-memory databases. With 32 cores per processor and 341GB/s of memory bandwidth to feed the processor cores with data from memory, the server excels in computational performance for analytics.

Making the move to HPE ProLiant DL385 Gen10 ensures your business is built on a solid infrastructure with the right memory architecture to support your data needs and enhance your memory-centric workloads. Curious to learn more? See what else HPE Gen10 servers can do for your enterprise.

Featured articles:


0 Kudos
About the Author


Ryan Wheeler is a data science leader with experience in some of the most cutting-edge technologies and trends. Wheeler has hands-on experience with machine learning, advanced analytics, and the largest big data systems in the world. From luxury fashion to social media companies in Silicon Valley, Wheeler is known for both his technical expertise and his ability to distill complex ideas clearly to C-level execs. Wheeler also writes code and predictive models on weekends, and whitepapers and blog posts on long plane flights