September 8, 2016 | Written by: Dylan Boday
Categorized: Power servers | Power Systems
Share this post:
Disrupt your industry without disrupting your data center
Today’s digital economy is demanding and growing exponentially. This growth is driven by the explosion of data coming from social media, IoT, AR/VR and AI. All are fundamentally changing the way businesses operate and compete. The ability to harness this data from a broad number of sources and draw real-time insights will lead to competitive advantages in areas such as fraud prevention, precision marketing and supply chain optimization.
When it comes to processing and capturing the value of today’s data growth, companies need to look beyond stagnant infrastructure. In the past, the Linux server market offered little choice and was built primarily on x86 processors designed within a monolithic ecosystem. In order to drive business innovation and industry disruption companies need a powerful alternative — a server that is designed to deliver superior data center efficiency and performance. IBM’s OpenPOWER LC servers are different by design. They are engineered to be more powerful and flexible than typical x86-based servers while leveraging industry-leading technologies from the most innovative companies within the OpenPOWER Foundation, to provide for the processing needs of tomorrow.
However, being different doesn’t mean IBM servers with OpenPOWER technology don’t fit in. These servers are designed to integrate seamlessly into your existing environments alongside existing infrastructure. Not only are these built to integrate with your existing data center and cloud environments, they deliver superior price-performance. Additionally, this server line is set apart with every server supporting innovations from the POWERAccel family of acceleration technologies. A great example of this dedication to acceleration can be observed with innovations such as our new POWER8 with NVlink processor, an industry first in processors. This technology enables the processor and the GPU accelerator to work in tandem with each other, and at much higher speeds, so that companies can experience acceleration the way it was intended.
Today, IBM is bringing three new OpenPOWER LC servers to the market. These servers provide the performance you need for your most demanding, data-intensive workloads, and are price-advantaged over comparatively configured Intel x86-based servers, costing 30 percent less in some configurations.
- IBM Power System S822LC for High Performance Computing: Clients can gain unprecedented performance and application gains with the new POWER8 with NVLink processor—delivering 5 times the CPU-GPU data flow compared to x86-based systems. These are the first servers in the industry to use NVLink for bandwidth improvements removing the traditional PCIe system bottleneck.
- IBM Power System S822LC for Big Data: designed to deliver differentiated big data performance and superior data through-put for Linux workloads. Customers can expect an average of 80 percent better price-performance on open source databases. Additionally, to handle the massive amounts of data today, this system provides remarkable virtualization density with 42 percent more VM’s per server than Intel Xeon E5-2690 v4 systems allowing you to contain server sprawl for the workloads compared.
- IBM Power System S821LC: Featuring a dense 1U form factor, this server packs the punch for your most compute-intensive workloads. Two POWER8 processors can now be accessed in a 1U server design. With this design, customers can gain more than 1.8 times more threads per 1U server compared to Xeon E5-2600 v4 systems. This system offers higher container densities and is ideal for dense database and virtualization workloads requiring a balanced system design of throughput and density.
Online pricing begins at $5999 dollars. Additional models with smaller configurations and lower pricing are available through IBM Business Partners.
IBM’s OpenPOWER LC servers were created with the most demanding business challenges in mind, helping businesses process data at the speed required by today’s digital economy. To learn more visit https://ibm.biz/Power-Systems-LC.
-  Results are based IBM Internal Measurements running CUDA H2D Bandwidth Test. Power System S822LC for HPC; 20 cores (2 x 10c chips) / 160 threads, POWER8; 2.9 GHz, 256 GB memory, 2 x 1TB SATA 7.2K rpm HDD, 2-port 10 GbEth, NVIDIA Tesla P100 GPU; Ubuntu 16.04. Competitive stack: 2x Xeon E5-2640 v4; 20 cores (2 x 10 chips) / 40 threads; Xeon E5-2640 v4; 2.4 GHz; 256 GB memory, 1 x 2TB SATA 7.2K rpm HDD, 2-port 10 GbEth, Device 0 of a Tesla K80 GPU (1 of 2 GK210 chips), Ubuntu 16.04
-  80% price-performance advantage is based on the average of IBM internal measurements of Power System S822LC for Big Data relative to comparable x86 E5-2690v4 (Broadwell) 2-socket offerings across multiple open source databases including MongoDB, EnterpriseDB, and MariaDB. Comparisons utilize current pricing as of August 24, 2016. More details can be found at https://www.ibm.com/developerworks/linux/perfcol/index.html
-  Results are based on IBM internal testing of single system running multiple virtual machines with Sysbench read only work load and are current as of August 22, 2016. Performance figures are based on running 24 M record scale factor per VM. Individual results will vary depending on individual workloads, configurations and conditions. IBM Power System S822LC for Big Data; 20 cores / 160 threads, POWER8; 2.9 GHz , 384GB memory MariaDB 10.1.16, 20 8vcpu VMs of Ubuntu 16.04 with KVM compared to competitive stack: HP Proliant DL380 28 cores / 56 threads; Intel E5-2690 v4, 6 GHz; 256 GB memory, MariaDB 10.1.16, 14 4 vcpu VMs of Ubuntu 16.04. with KVM.Each system was configured to run at similar per VM throughput levels and number of VMs were increased for each system until total system throughput showed maximum throughput levels. Competitive pricing was taken from available web-based pricing. For more information about MariaDB go to: http://mariadb.org/. Pricing is based on: S822LC for Big Data https://www.ibm.com/it-infrastructure/power/os/linux and HP DL380 https://h22174.www2.hp.com/SimplifiedConfig/Index