Albireo VDO

The Albireo Virtual Data Optimizer (VDO) is a drop-in data efficiency solution built for Linux architectures. It provides fine-grained (4k chunk), inline deduplication , thin provisioning, compression and replication. VDO leverages existing file systems, volume management and data protection, providing the fastest route to market for delivering data efficiency solutions in a Linux storage environment.

Key Features

  • Out of the box data efficiency for Linux storage solutions
  • Performance >4GB/s throughput
  • Most resource efficient solution in the industry
  • Scales to 256/TB
  • Optimized for Flash and HDD environments
  • Unique, synchronized compression with deduplication provides highest performance in storage industry
  • Replication combined with deduplication provides the most efficient method for network transfer of duplicate data on the market today

VDO Features and Benefits

Feature Benefit
Out-of-the-box data efficiency for Linux storage solutions
  • Take advantage of Linux cost savings and integrate a drop-in data efficiency solution with no additional development to gain 5-35x data reduction savings
  • Accelerate time to market
  • Differentiate your product and deliver competitive advantage while increasing revenue
  • Implement in weeks; get to market in months
Industry Leading performance
  • Lower effective cost of all storage (flash, hybrid and disk) by 5x-35x.
  • Reduce effective cost of flash comparable to HDD
  • Implement on highest performing arrays – Inline efficiency cuts writes and extends media life 5-35x extending life of flash cells
  • Eliminate data cache inefficiency via inline performance
Extreme Resource Efficiency
  • Reduce costs by lowering system BOM
  • Retro-fit current platforms, enables in-field upgrades, extending life cycle of products
  • Increase margins and retain customer base
Scalability
  • Scalability that addresses high data growth requirements,  enabling SLA attainment
  • Avoid LUN and volume size management expenses
  • Inline single pass dedupe prevents buffering / writing to disk, increasing operating efficiency
Optimized for Flash and HDD
  • Eliminate data inflation between flash and HDD by delivering highest performance for both HDD and flash-based systems with our optimized hashing algorithms
Smart Compression (optional)
  • Further reduce storage and operating costs using Smart Compress, which dedupes then compresses, delivering the highest levels of data reduction and performance
Replication (optional)
  • Reduce network consumption and costs with industry leading bandwidth performance

Business Impact

By 2016 all primary enterprise storage systems will incorporate data efficiency technologies. With Albireo VDO, OEMs can dramatically reduce their time to market in delivering data deduplication and thin provisioning with optional compression and replication within their existing Linux-based product offerings. By leveraging existing Linux storage infrastructures, Albireo provides the ability to deliver the most scalable, highest performing and resource efficient data efficiency technology in the industry.  By adding these efficiency technologies you will be able to maintain account control, increase profit margins, and take away market share from your competitors. Either you can deliver this to your customers or your competitors will. Don’t waste years trying to build deduplication from scratch. Permabit has been delivering this technology since 2009. Let us help you accelerate your time to market reducing it from years down to weeks!

How It Works

The Albireo VDO Linux kernel module provides deduplication and thin provisioning, and is exported as a normal block device that can be used directly as block storage or presented through one of the many available Linux file systems (such as ext3 and XFS).

As a block device, VDO sees write requests that come down from the file system. VDO takes the new blocks and uses the Albireo deduplication technology to find duplicate content. VDO then manages the writing of any unique blocks to the storage device. VDO keeps track of stored data and manages the data in a way that is seamless to the user. Data optimization through block-level deduplication increases the overall capacity of the underlying device.

VDO-Architecture
  1. VDO kernel module receives request to write a block of data to a VDO volume
  2. Data not otherwise part of a special case are captured by VDO
  3. Unique fingerprints computed
  4. Patented indexing technologies determine if chunk has been previously seen
  5. Byte-by-byte comparison of candidate duplicate block is computed
  6. Any matching blocks are deduplicated and all unique blocks are stored and both cases are noted by the index

Albireo VDO can be extended with the following data management enhancements:

Albireo COMPRESS

Albireo COMPRESS provides a plug-in solution for Linux-based storage OEMs that utilize VDO. Like VDO, COMPRESS supports all major Linux versions used by the storage industry.

COMPRESS and VDO work together in a unique, synchronized fashion. While deduplication works at a “macro” level (with typical reduction rates of 4-35x), compression works at the “micro” level (with typical rates of 2-4x).

Deduplication is the first step, with VDO analyzing data to determine if it has previously been seen. Compression is then applied to the remaining unique data. Using deduplication as the first step ensures that redundant copes are dealt with efficiently, without requiring an additional compression step, thus saving CPU cycles.

VDO and Compress work in tandem: Compression identifies “micro” duplicates (2–4 X) – Dedupe identifies “macro” duplicates (4-100 X)

Albireo REPLICA

Albireo REPLICA is a remote file copy service that works in concert with Albireo VDO to provide data optimization for content both over the wire and at rest. REPLICA software optimizes data before it is copied to a VDO-enabled device running the REPLICA service. The result is industry-leading bandwidth optimization that surpasses the capabilities of other products on the market today.

Key Features Explained

Performance
VDO delivers scalable deduplication throughput performance of up to 4 GB/second and up to 256TB capacity. It dramatically improves native write performance by eliminating duplicate data, and utilizes fewer hardware resources than any other solution available today.

Extreme Resource Efficiency
Performance and scalability do not come at the expense of massive amounts of RAM consumption. VDO is the most resource efficient solution in the industry. It can operate on a single x86-64 CPU core, with as low as 0.11 GB/TB of DRAM and 16 GB/TB of disk. This allows OEMs to deliver lower system bill of materials and larger array size. See Table 1 for select Albireo memory utilization metrics.

Table 1: Select Albireo Metrics (* Assumes 10x logical storage)

System Capacity 1.0 TB 16 TB 64 TB 256 TB
Total Memory Reqs 0.5 GB 1.9 GB 7 GB 28 GB
Disk Reqs* 16 GB 263 GB 1.0 TB 4.2 TB

Scalability
The Albireo VDO scales to 256/TB of raw capacity for an easy to integrate, purpose built OEM inline deduplication solution. This is an order of magnitude greater than the largest flash array system with built-in deduplication available.

Product VDO Enabled
Storage
EMC XtremIO PureStorage
FlashArray
Raw Capacity 256 TB 20 TB 35 TB

Optimized for Flash and HDD Environments
Supports synchronous and asynchronous write paths

Smart Compression (optional)
Synchronized deduplication and compression to provide highest performance in storage industry.

Optional Albireo COMPRESS seamlessly plugs in to VDO to compress data after it has been deduplicated minimizing resource requirements.

Replication (optional)
Albireo REPLICA is a remote file copy service that works in concert with Albireo VDO to provide data optimization for content both over the wire and at rest. REPLICA software optimizes data before it is copied to a VDO-enabled device running the REPLICA service. The result is industry-leading bandwidth optimization that surpasses the capabilities of other products on the market today.

Rapid Time to Market
VDO is a complete turn-key solution for Linux environments that supports existing Linux-based NAS, SAN, and unified storage solutions without modification and can be easily integrated into existing storage solutions in weeks and deployed within months.

Albireo VDO is right for you if…

  • You need to deliver advanced storage features like deduplication, compression, thin provisioning and bandwidth optimized replication on Linux
  • You already leverage popular components of the Linux block storage stack such as md and LVM2
  • You want to be able to integrate existing software into your solution in weeks