Servers are the heart of data centers, which means they can make or break a major initiative, such as migrating to a software-defined architecture or supporting high-performance computing. If your organization is gearing up for that kind of big change, consider upgrading to servers based on Intel Xeon Scalable processor architecture.
The latest generation of these servers is part of Intel’s overarching software-defined infrastructure evolution. For starters, the Xeon Scalable boosts data center performance through higher frequencies and an increase in processor cores. The table below summarizes how each of the four server segments benefits from the SDI environment that Xeon processors help enable.
|Server Segment||Server Description||Workload||Capacity|
|Basic||Single-socket micro- or low-frequency||For cost-sensitive workloads, such as infrastructure and web servers, that require low core count and low performance, and operate on a smaller data size.||<1%|
|Performance||80%: Higher frequency, single socket,
32 GB for workloads within 8 GB per core
20%: Higher frequency, 32 to 512 GB
|For most workloads that require faster application responses, better data availability and proximity, and low-latency data access. This workload arrangement reduces licensing costs and lowers energy consumption||>78%|
|Throughput||Dual socket with 10 or more cores
|For workloads that need to maximize the number of jobs to be completed within a fixed time frame.||20%|
|Massive memory||Four socket, up to 6 TB RAM per serve||For high-memory workloads, including Design mask and tapeout applications that handle complex design data, Office and Enterprise applications that handle complex data warehousing, and financial data that requires quicker processing.||<1%|
Table 1. Four server segments for software-defined compute SOURCE: Intel
This generation also features a new memory architecture that provides up to 6-terabyte capacity with support for persistent data. That’s ideal for Internet of Things (IoT) data and in-memory database applications, such as Oracle and SAP HANA.
A variety of new features also help protect data, including the flood of data you can expect from IoT endpoints and mobile devices. To be inexpensive, IoT devices typically have just enough processing power and memory for their core tasks, with little left over for security tools such as malware detection. As a result, the data center now shoulders more of the security workload. That’s where features such as Intel Trusted Execution Technology (TXT) come in handy.
- Verified launch: a hardware-based chain of trust that enables fallback to a “known good” state after the system detects changes indicating a security breach
- Launch control: a policy engine for creating and implementing enforceable lists of known good or approved executable code
- Secret protection: a set of hardware-assisted methods that remove residual data when the measured launch environment shuts down improperly and thus protects data from memory-snooping software and reset attacks
- Attestation: provides platform measurement credentials to local or remote users or systems as part of the trust verification process, along with supporting compliance and audit processes
The new Xeon-based servers also have compression, offload and other features that enable more efficient networking, a major plus for Big Data workloads. Some of the gains include:
- Optional integrated hardware accelerators help with network security, routing, storage and Big Data. Benefits include maximum CPU utilization, faster secure tunnels and the ability to support more authenticated clients.
- Intel QuickAssist Technology enables high performance of encrypted traffic across the secured network. It also accelerates compute-intense symmetric and asymmetric cryptography.
- The compressed file system data blocks enable quick analytics, including faster Hadoop run times with Big Data workloads.
- Compressing data in real time enables high-performance storage.
If you’re looking to migrate to an SDI environment, the Intel Xeon Scalable-based servers belong on your short list. They’re ideal for providing the efficiency, security and performance that demanding applications such as Big Data require, and they’ll give you a solid foundation for years of data center growth.
This blog post brought to you by: