Peek Inside FlashBlade™

Have a closer look at three key components: Blade, Purity, and Fabric.

Our Four Design Principles

True Scale-Out

FlashBlade delivers performance that grows linearly with more data. It is built to scale to 10s of PBs and 10s of billions of files/objects. An elastic scale-out design at every level, FlashBlade is simple to manage at any size. Need anything more? Just add a blade.

Effortless to Manage

FlashBlade includes a built-in GUI and CLI, as well as our Pure1 cloud-based management and support platform. A single administrator can manage a FlashBlade deployment of any size.

Enterprise-Class Reliability

With an N+2 design for redundancy, FlashBlade can tolerate two full blade failures and many more flash chip failures without data service interruption. The latest LDPC error recovery protects against flash bit errors, and all data is encrypted all the time. Blade loss results in a predictable 1/N aggregate performance loss, and the system auto-rebuilds to heal around failures.

Integrated, Software-Defined Fabric

We designed intelligence into the networking so you don’t have to worry about it. FlashBlade utilizes a high performance 320Gb/s fabric with object messaging and distributed transaction protocol to ensure fast and reliable communication between Purity instances running on each blade.

EACH BLADE ADDS MORE MAGIC

The simple scaling unit that adds DirectFlash™ capacity (8TB, 17TB, or 52TB), along with protocol processing, metadata, and more compute cores with each addition.

Enterprise-Class
Xeon CPU

Up to 8 CPU cores + networking in a high performance design

Programmable Processor

FPGA with embedded ARM cores delivers efficient flash management

PCIe Connectivity

The fastest connection from flash to processors

DRAM with Power-Safe Write Protection

Scales write bandwidth with each Blade


Purity for flashblade

Powering scale-out storage services at every layer, with DirectFlash technology for global flash management.

Core Services

As the architectural foundation of the system, Core Services offers a scale-out metadata design powered by a massively distributed transaction database. A variable-block metadata engine delivers the highest performance regardless of data type or size. Scale-out fabric with DirectFlash delivers the ultra-low latency, high bandwidth performance that only a purpose-built system can deliver.

Data Services

Purity Data Services are based on variable-block metadata across all layers of the system. Always-on compression, thin provisioning, encryption, and rapid data locking, as well as HA and NDU, are built-in. Purity provides global flash management via DirectFlash. As the latest addition to Purity//FB, snapshots offer customers the highest level of protection for enterprises and software development use cases. 

Protocols

Built for unstructured data, Purity//FB supports all major protocols natively on FlashBlade, such as NFS, SMB, S3/Object, and HTTP. Each protocol is implemented for true scale-out deployment – built on the distributed transaction database in Core Services to deliver unprecedented performance at any scale. The modular design of Core Services allows for extensibility to future protocols.

SCALE-OUT FABRIC

A built-in, software-defined fabric that scales to 100s of Blades and 1000s of clients.

High Performance 320Gb/s

From client to Blade, a 320Gb/s Ethernet network ensures the fastest, most cost-effective connectivity that you never have to manage.

Proprietary Low-Latency Protocols

Proprietary object messaging and distributed transaction protocols ensure fast and reliable communication between Purity instances running on each blade.

Software-Defined
Fabric

Separate virtual networks isolate client, management/metadata, and data transport with defined SLAs.