The 10 Hottest New Business And Enterprise Servers Of 2019

With the worldwide server market poised to exceed $100 billion in sales this year, CRN breaks down the ten business and enterprise servers that led the way in 2019.

The demand for enterprise computing over the past two years has been at an historic high, as businesses seek next-generation server performance to meet their digital transformation needs.

In 2019, the worldwide server market is projected to exceed $100 billion in revenue with each quarter generating on average more than $20 billion in server sales, according to IT market research firm IDC. Although market leaders like Dell Technologies and Hewlett Packard Enterprise witnessed a fall in server sales this year, the global server market is expected to pick back up in 2020.

Hardware server innovation is being led by the likes of Dell, HPE, Inspur and Lenovo who are accelerating computing performance, storage class memory and next-generation input/output (I/O) for workloads around artificial intelligence, cloud, virtual desktop infrastructure (VDI) and edge computing.

CRN breaks down the ten hottest business and enterprise servers that led the way in 2019.

Get more of CRN's 2019 tech year in review.

Cisco C480 M5 Rack Server

Included in Cisco’s flagship hyperconverged infrastructure platform HyperFlex, the company’s UCS C480 M5 rack server delivers industry leading performance for in-memory databases, big data analytics, virtualization, Virtual Desktop Infrastructure (VDI) and bare-metal applications. The 4U form factor includes four Intel Xeon Scalable processors with up to 28 cores per socket with options for NVIDIA and AMD. The C480 M5 has support for the Intel Optane DC Persistent Memory with up to 512G as well as DDR4 memory from 6TB to 12TB. The UCS server has 12 PCIe slots, Cisco’s 12-Gbps SAS Modular RAID Controller and dual embedded 10 Gigabit Ethernet ports. The C480 has flexible storage options with support for up to 32 small-form-factor 2.5-inch, SAS, SATA and PCIe disk drives.

Dell Technologies PowerEdge MX840c

Designed for Dell Technologies’ PowerEdge MX kinetic infrastructure and its new all-in-one PowerOne hardware stack, the company’s PowerEdge MX840c server is storage rich with flexible 4-socket compute sled ideal for database-driven mission critical applications, performance workloads and in dense virtualization environments. The MX840c has been designed for easy access to components for efficient cooling and supports up to two processors, 48 DDR4 DIMM slots, four mezzanine cards, and two mini-mezzanine cards. Dell’s PowerEdge servers features built-in security, embedded agent-free integrated Dell Remote Access Controller (iDRAC) 9 with Lifecycle Controller to simplify server systems management. Last month, Dell unveiled new GPU and Intel FPGA support for its PowerEdge servers including NVIDIA Tesla V100S GPU and PowerEdge Express Flash NVMe Performance PCIe SSDs.

Dell Technologies DSS 8440

At Dell Technologies World this year, the Round Rock, Texas-based launched the DSS 8440 accelerator-optimized server, specifically designed for high performance machine learning training. With a choice of 4, 8 or 10 of NVIDIA’s Tesla V100 Tensor Core GPUs -- combined with 2 Intel CPUs for system functions, a high performance PCIe fabric for rapid IO and up to 10 local NVMe and SATA drives for optimized access to data -- the DSS 8440 server has both flexibility and top level performance. Dell’s DSS 8440 is an ideal solution for machine learning training as well as other compute-intensive workloads like simulation, modeling and predictive analysis in engineering and scientific environments.

HPE ProLiant DL325

Equipped with second generation AMD EPYC processors, HPE’s ProLiant DL325 delivered an 321 percent virtualization performance boost over the previous single socket server record holder this year. The DL325 has up to 64 cores, 16 DIMMs, and 4 TB of memory capacity, as well as support for up to 24 NVMe drives. Customers get increased data-centric application performance with HPE’s DDR4 SmartMemory running at memory speeds up to 3200 MT/s. The DL325 provides an agility advantage in database provisioning and maintenance for cloud, big data and IoT environments. In terms of security, the server includes built-in innovative firmware protection, malware detection and firmware recovery capabilities right down to the silicon.

HPE ProLiant DL385

The HPE ProLiant DL385 with two second generation Gen AMD EPYC processors crushed the previous virtualization world record this year with 61 percent better virtualization performance at a 29 percent better price-performance ratio. HPE said the DL385 is the world’s most power efficient two-processor server. The ProLiant DL385 has up to 128 cores, 32 DIMMs, and 4 TB of memory capacity per CPU, as well as 32 NVMe drives. The HPE ProLiant DL385 Gen10 server specializes in targeting high-performance computing (HPC) applications to address complex and compute-intensive workloads. The server is also part of HPE’s GreenLake on-premises hybrid cloud consumption-based as-a-service solution.

IBM Power System E980

IBM’s most powerful and scalable server, the Power System E980 is a high-end enterprise server that provides rapid scalability, enhanced security and simplified management. An E980 server can have 4, 8, 12 or 16 sockets, and up to 192 IBM POWER9 cores. A four-node system can have up to 64 TB of memory with bandwidth of 920GB. Each system node has 32 customer DIMM slots. The processor has EnergyScale technology with variable processor frequency modes as well as support of up to four NVMe SSDs in each node. IBM’s Power System E980 delivers industry-leading virtualization, capacity on demand for processors and memory, and flexibility for multiple systems in an enterprise private cloud.

Inspur NF8380M5

Inspur’s NF8380M5 four-socket server is designed for AI training, efficient data warehousing applications and hyperscale cloud workloads. With up to four Intel Xeon processors, the NF8380M5 is a 3U based on the Project Olympus four-socket server developed by Intel Purley Platform. The server has 48 DDR4 DIMM slots, up to 6TB of memory and was recently accepted to the Open Compute Project (OCP). The NF8380M5 was designed for multi-scene AI applications and can drive GPU acceleration training workloads when serving as compute head node. Inspur’s server supports all NVMe SSDs configuration for real-time decision database applications.

Inspur NE5250M5

Inspur, which ranks third in worldwide server market share, launched a new artificial intelligence server for edge computing applications this year with the Inspur NE5250M5. Featuring two NVIDIA V100 Tensor Core GPUs or six NVIDIA T4 GPUs, the new Inspur server is ideal for compute-intensive AI applications, including autonomous vehicles and smart cities. The NE5250M5 features a telecom 2U server form factor and supports remote online maintenance and management, which includes a self-repair mechanism where it automatically returns to pre-maintenance design. The server aims at enabling 5G edge applications and supports harsh deployment environments with the ability to withstand electromagnetism, heat and corrosion.

Lenovo ThinkSystem SE530

Recently validated for Microsoft’s Azure Stack HCI, Lenovo’s ThinkSystem SE530 server is purpose-built for the edge. The high-performance two-socket server features Intel’s Xeon processors, zero-touch deployment software and is managed with Lenovo XClarity Controllers. The ThinkSystem SE530 includes 256GB of RAM, up to 16 DIMMs, 16 TB of internal solid-state storage and two PCIe slots. The server comes with built-in encrypted storage and physical security such as tamper-detection mechanisms. The SE530 provides intelligent and adaptive system performance with Intel’s Turbo Boost Technology 2.0 that allows CPU cores to run at maximum speeds during peak workloads by temporarily going beyond processor thermal design power.

Lenovo ThinkSystem SR670

Hong Kong-based Lenovo, which currently ranks fourth in worldwide server market share, designed its 2U ThinkSystem SR670 rack server to support up to four large or eight small GPUs per node. Built on Intel Xeon Scalable processors, the ThinkSystem SR670 is ideal for running artificial intelligence, HPC and VDI workloads with a ratio of up to 1:4 for CPUs to GPUs. The server includes an XClarity Controller (XCC) to monitor server availability, Single Device Data Correction, memory mirroring, and memory rank sparing for redundancy in the event of a failure, as well as energy-efficient planar components to help lower operational costs.