Five of the biggest systems vendors – Dell EMC, Lenovo, Supermicro, Cisco and Cray – in concert with Intel’s launch this morning of its second-generation Xeon Scalable processors (Cascade Lake) and Optane persistent memory – announced the refresh of server portfolios leveraging the new Intel technologies.

Here’s a summary of the server companies’ announcements:

Dell EMC announced upgrades to its PowerEdge server line with new management, security, performance and scalability capabilities, according to the company, along with adoption of the new Intel Xeon Scalable processors and Intel Optane DC persistent memory for improved workload performance of up to 2.7 times more transactions per second, according to Dell EMC.

Key features, according to Dell EMC, include:

• Management: Enhancements to the Redfish-compliant Dell EMC RESTful API for control of server operations and devices, including management of servers from an operating system in addition to a typical management network, enabling simplified integration of PowerEdge servers into existing processes. Also, the servers offer augmented reality on OpenManage Mobile, giving a graphical representation on mobile devices of where incidents occur within PowerEdge MX infrastructure.
• Security: Dell EMC said its OpenManage Secure Enterprise Key Manager scales data protection and streamlines management of self-encrypting drives, enabling central management for self-encrypting drives to better secure data at the drive and server level.
• Performance: The company said the new Intel processors offer up to a 40 percent increase in performance. According to a new internal benchmark, a Dell EMC PowerEdge R940 server with the new Xeon chips can host 69,500 simultaneous users within the SAP Sales & Distribution Standard Application Benchmark, up to a 60 percent increase.
• With new Intel Optane DC persistent memory, workloads can be accelerated with up to 2.5 times more memory capacity, according to the company. The PowerEdge R740xd enables up to 2.7 times the transactions per second with an Optane DC persistent memory compared to NVMe drives in a virtualized Microsoft SQL Server 2019 preview environment with VMware ESXi, the company said.

Dell EMC PowerEdge R940, R740, R740xd, R740xd2, R640, and C6420 servers with the new Xeon processors will be available by the middle of this month, the company said. Additional PowerEdge servers, including those for PowerEdge MX, will be globally available with the processor upgrade in the second quarter of 2019.

Cray Inc. announced product line updates to support Intel Xeon Gold 6200 and Xeon Platinum 8200 processors in both its CS500 and XC50 supercomputers. Cray has also confirmed it will support the Xeon Platinum 9200 (Cascade Lake-AP) SKUs on both its CS500 systems and coming Shasta supercomputers, set to debut in 2020.

“The market is entering a new era of exascale-class computing where customers will increasingly use a combination of HPC and AI methodologies to aid in their work and research. To power this new era, customers will have an ever-increasing need for the latest in processing capabilities optimized for this convergence,” said Fred Kohout, senior vice president and CMO at Cray. “The added support for the new 2nd gen Intel Xeon Scalable processors in our XC50 and CS500 supercomputers provides customers with the performance, scalability and flexibility to address their rapidly evolving analytics needs to achieve the fastest time to insight.”

Lenovo announced 15 new ThinkSystem servers and five ThinkAgile appliances powered by the new Xeon chips, including what the company said is the industry’s first eight-socket server with Optane technology, providing 12.5 times faster recovery from outages and “can process significantly more data in in-memory databases, making the solutions ideal for SAP HANA environments.”

In addition, Lenovo said it is developing engineered solutions for SAP HANA, Microsoft SQL Server and Red Hat OpenShift Containers workloads that will be verified by Intel as Intel Select Solutions.

Supermicro announced its entire X11 portfolio has adopted the new Xeon processors, Optane and Intel Deep Learning Boost for more efficient AI acceleration. The company said “resource-saving servers, (deliver) better data center performance (35 percent faster), better TCO (up to 50 percent reduction) and less impact on the environment.”

The company said its 1U and 2U Ultra SuperServers feature dual Xeon processors (up to 28 cores, 205W TDP per CPU), 24 DIMMs of DDR4-2933MHz Reg. ECC memory, Intel Optane DC persistent memory, up to 24 hot-swap NVMe (up to 16 million IOPS) or SAS3/SATA3 drives, up to eight PCI-E 3.0 expansion slots, dual or quad-port 1G, 10GBase-T, 10G SFP+, or 25G SFP28 Ethernet, and redundant 750W/1000W/1600W Titanium Level (96 percent-plus) digital power supplies.

Supermicro also announced that its new petascale line of all-flash NVMe 1U storage servers support next-generation flash technologies with up to 52GB/second data bandwidth. The company said its 1U systems support up to 1PB of storage with 32 front hot-swap U.2, EDSFF and NF1 form factor SSDs.

Cisco announced its UCS servers also have been updated with 2nd Generation Xeon Scalable processors and with Optane memory with 128GB, 256GB and 512GB module capacities.

“Intel Optane DC persistent memory is a new technology designed to fill the capacity, cost, and performance gaps between traditional DRAM memory and storage for servers,” stated Cisco’s Bill Shields, senior marketing manager, Product and Solutions Marketing Team, in a blog today. “While the performance of DRAM is great, it is relatively expensive and volatile (the contents disappear when the server is rebooted). SSDs, one option for storing programs and data, while faster than hard disk drives, are not nearly as fast as DIMMs, though their content isn’t volatile.”

Optane was designed to provide an option between DRAM and SSD, Shields said, “reducing the cost while potentially increasing the size of server memory and/or providing the fastest persistent storage of data.”

He said Cisco’s UCS Manager can dynamically define how customers use Optane, in memory or app direct mode, or both.

When in memory mode, Optane modules are paired with DDR4 DIMMs to cut memory cost, Shields said. There are two options to increase per-socket memory to 3 TB: 12x 256GB DDR4 DIMMs, or the more cost-effective option of 6x DDR4 DIMMs + 6x 512GB Intel Optane DC persistent memory modules (the DDR4 DIMMs are “invisible” to the operating system and instead act as a fast cache for the persistent memory modules).

In app direct mode, Optane modules are used as low-latency, high-speed persistent memory for data or storage attached directly to the CPU instead of the PCIe bus. Each CPU can have up to 1.5 TB of DDR4 memory and up to 3 TB of App Direct memory / storage.

Other announced launch partners include: Huawei, Inspur, Tyan, RSC, Colfax, Penguin Computing, Advanced Clustering Technologies, Wiwynn, NetApp – as well as Google, Amazon, Baidu, Alibaba and XTREME-D.