BeeGFS


BeeGFS (formerly FhGFS) is the leading parallel cluster file system, developed with a strong focus on performance and designed for very easy installation and management. It is based on a lightweight architecture and can be created on a per job basis with the BeeOND (BeeGFS On Demand) application. This BeeOND feature is designed to provide new instances on the fly across all compute nodes being assigned to a particular job. BeeOND aggregates the performance and capacity of internal SSDs or hard disks in compute nodes for the duration of a compute job to increase performance levels with a very elegant method of burst buffering, which can be highly useful in cloud environments and temporary scratch data scenarios. If I/O intensive workloads are your problem, BeeGFS is the solution.
Image
BeeGFS file system’s performance and capacity can seamlessly be scaled out from small clusters to enterprise-class systems with thousands of nodes. RAID Inc. supports BeeGFS and builds storage solutions using BeeGFS to match customer needs. BeeGFS® was optimized especially for use in HPC:

Best in class client throughput (3.5 GB/s writes and 4.0 GB/s reads with a single I/O stream on FDR Infiniband)

Best in class metadata performance: linear scalability through dynamic metadata namespace partitioning

Best in class storage throughput: BeeGFS® servers allow flexible choice of underlying file system to perfectly fit the given storage hardware.

With performance-centric highly available data increasing in demand, a flexible architecture like BeeGFS eliminates data silos and storage complexity with a solution built specifically for multi-tenancy and cloud. BeeGFS is a parallel file system – by adding more servers, the capacity and performance of them is aggregated in a single namespace. That way the filesystem performance and capacity can be scaled to the level which is required for the specific application. Furthermore, BeeGFS is splitting MetaData from ObjectData. The ObjectData is the data users want to store, whereas the MetaData is the “data about data”, such as access rights and file size – but most important in the MetaData is the information, on which of the numerous storage servers the content of a file can be found.

Image
Image
The BeeGFS parallel file system was developed for performance-critical environments, with a focus on easy installation and high flexibility. It supports a wide range of Linux distributions, including RHEL/Fedora, SLES/OpenSuse and Debian/Ubuntu and a wide range of Linux kernels.

Storage servers run on top of existing local filesystems. Clients and servers can be added to an existing system without any downtime.

Multiple networks and dynamic failover are supported for when network connections are down.

Multiple services can run on the same physical machine. This eliminates the need for external storage resources, providing a cost-efficient solution with simplified management.

BeeGFS is based on a lightweight architecture and can be created on a per job basis with the BeeOND (BeeGFS On Demand) application. BeeOND aggregates the performance and capacity of internal SSDs or hard disks in compute nodes for the duration of a compute job to increase performance levels with a very elegant method of burst buffering, which can be highly useful in cloud environments and temporary scratch data scenarios.

The BeeGFS® servers are user space daemons and do not require special administration rights.

BeeGFS® also was designed with easy administration in mind. The GUI enables easy user management: cluster installation, storage service management, live throughput and health monitoring, file browsing, striping configuration, and more.


Image

KEY FEATURES

Data and metadata for each file are separated into stripes

Secure authentication process between client and server to protect sensitive data

Fair I/O option on user level to prevent a single user with multiple requests to stall other user requests

Storage high-availability based on replication

Automatic network failover

Runs on non-standard architectures

Optional server failover for external shared storage devices

Want to know how we can optimize BeeGFS file-system for HPC?
Talk to an Expert

RAID Inc. Storage Systems with BeeGFS


Enterprises are focused on addressing challenges with data that is expanding at a faster rate like never before. It has become a challenge for industries to perform data utilization at speeds that help faster innovation. Those that can deliver insights faster while managing rapid infrastructure growth are the leaders in their industry. An organization’s underlying storage must support both new-era big data and traditional applications with security, reliability and high-performance. RAID Inc storage systems optimized for BeeGFS will meet your specific HPC needs and challenges. Our high-performance file-system solutions help you manage your data at scale with the distinctive ability to perform IO with faster speeds while being reliable along with options like archiving and analytics in place.

Gain performance with BeeGFS and RAID Inc. storage.
Request A Quote

Case Studies & Other Resources



So many satisfied
Customers

Our Approach


To deliver our tailored HPC solutions, RAID Inc.'s consultative sales and system design process is customized every step of the way. From the initial sales call to installation day, RAID Inc. works closely with clients to assess their unique computing and storage requirements by examining every facet of their computing needs. To learn more about our
6-Step Holistic Approach, 
click here.

Find

RAID Inc. performs rigorous requirements engineering strategies to develop solutions based on the actual need. It incorporates strategic approaches to find and understand your IO performance requirements when deploying systems and storage while keeping your costs low.

Analyze

RAID Inc. analyzes the best systems and storage solutions based on the requirements elicitation performed. With a wide range of compute and networking products available in the market, we deploy the right servers and networking products along with our storage to meet you actual needs.

Solve

RAID Inc. solves your HPC problems in no time. Being an industry leader, we served 1000s of customers in solving their HPC problems. The areas we covered include but not limited to genomic research, space explorations, weather forecasting, gas and oil explorations, governments, educational institutions, defense, life sciences and many others.