Skip to Content

Penguin IceBreaker Storage Servers

IceBreaker Storage

All IceBreakers come with a three-year hardware warranty and will be configured to run your Linux file system of choice.

Penguin Computing offers a wide array of storage solutions, from a single storage server to a complex, scalable high-performance parallel file system

Storage Servers

Storage Solutions

Storage Servers

Based on the latest generation of Intel Xeon E5-2600 processors, Penguin Computing’s IceBreakers offer scalability, flexibility and great price-performance. The IceBreaker family is available with 3.5’’ and 2.5’’ SAS and SATA hard drives or SSDs in a variety of capacities and performance levels.

  • IceBreakers can accommodate up to 36 hard drives with over 100TB of storage capacity in a 4U form factor.
  • IceBreaker servers and complementing JBOD storage arrays are the ideal solution for cost-effective bulk and nearline storage.
  • The PCI-E gen 3 bus architecture makes the IceBreaker a future-proof investment.
  • DDR3 memory speeds of up to 1600MHz and 4 memory channels allow for highly efficient data caching.
  • All IceBreakers can accommodate multiple network adapters and RAID controllers for data delivery at full speed.

With four chassis form factors and multiple drive options, the versatile IceBreaker can be configured to fit a variety of roles ranging from a simple NFS/CIFS file server to an object storage server in a sophisticated, high-performing distributed Lustre file system. For cost efficient nearline storage, Penguin offers a variety of external JBOD enclosures that deliver capacities up to 135TB in a 4U form factor.

High Availability Storage Arrays

For storage environments with stringent availability requirements Penguin’s Icebreaker HA product is the right choice. The Icebreaker HA features fully redundant RAID controllers with mirrored caches making it a great building block for highly available Lustre or NFS configurations. The Icebreaker HA is available with SAS storage connectors for direct attachment or with an FC connectivity option.

Solutions

UDX1 Storage Solution - Ceph

Ceph is a massively scalable, open source, software defined storage system. It comprises an object store, block store, and a POSIX- compatible distributed file system. The platform is capable of auto-scaling to the exabyte level and beyond, is self-healing and self-managing, and has no single point of failure. The UDX1-Ceph storage solution integrates the distributed, highly scalable Ceph storage system with Penguin Computing’s power efficient UDX1, a server system based on Calxeda’s ARM based Energy Core SoC

Resources

Icebreaker CS / Scality RING - Object Oriented Large Scale Cloud Storage

Scality RING is a new object-oriented software solution for storing massive amounts of unstructured data. Scality chose an object storage philosophy to address petabyte scale challenges such as fault tolerance, performance and manageability. Based on Penguin Computing's Icebreaker 4860 as a building block Penguin Computing and Scality have jointly developed a highly scalable storage appliance with a unique feature set and unmatched scalability.

Resources

Penguin Computing’s Lustre solution greatly complements our new 50TFLOP compute cluster that requires a highly scalable I/O solution in the back-end to deliver the expected levels of performance. Given Penguin’s track record, the positive experiences we have had with Penguin in the past and their ability to provide complete support for our cluster as well as our Lustre storage solution, it made perfect sense to choose Penguin for this deployment.

Daniel J. Grim, Chief Technology Officer, Information Technologies at the University of Delaware

Lustre - HPC Storage

For optimal I/O performance for HPC clusters, Penguin offers storage solutions based on the distributed high-performance file system, Lustre. Lustre provides a single file system and name space across multiple storage servers. With a distributed storage architecture, Lustre can be scaled in two dimensions: Capacity can be increased at the level of the individual storage server or through the addition of storage servers for improving performance as the number of Lustre clients grows. Due to superior scalability and performance, Lustre technology powers over 70 percent of the top 100 supercomputers and is used in some of the world’s largest commercial, university, research and government environments.

In partnership with Whamcloud, the driving force behind Lustre development, Penguin Computing offers fully supported Lustre solutions that are based on best-of-breed hardware components. Penguin’s Lustre solutions are pre-integrated in Penguin Computing’s factory and can be quickly moved to production right after delivery.

Download our data sheet to find out more about our Lustre solution.

Hadoop - Big Data Storage

Hadoop, a scalable and flexible open source software framework for the distributed processing of large data sets on compute clusters has rapidly emerged as the preferred solution for big-data analytics. Web giants that are dealing with large-scale clickstream analysis, search and targeting scenarios such as Yahoo!, LinkedIn, Facebook, e-Harmony and e-Bay, have all adopted Hadoop.

Hadoop

Penguin Hadoop Stack (click to view larger).

Penguin Computing has deployed hundreds of nodes in Hadoop environments. In cooperation with Cloudera, Penguin has designed a scalable Hadoop solution that includes all software functions and services to run Hadoop in a production environment including management tools for monitoring Hadoop, cluster load and individual servers.