edge-scale

Data Efficiency at Scale

| By: (56)

Data Efficiency at Scale In the past, I’ve written about the performance challenges associated with implementing data efficiency features (deduplication and compression). Today I’d like to spend some time on the related topic of scalability. The initial wave of data efficiency features for primary storage focus on silos of information organized in terms of individual file systems. Deduplication and compression features provided by some vendors are limited by the scalability…

Read more

Linux Penguin

Albireo VDO Makes Linux Scream

| By: (56)

Permabit Labs is at it again! This time testing the new Logical Volume Manager (LVM) snapshot capabilities provided by the dm_thin device mapper target for Linux. Permabit’s results show that when dm_thin snapshot capabilities are combined with Albireo Virtual Data Optimizer (VDO), the combined solution outperforms dm_thin alone. What are snapshots and how are they managed on Linux? Snapshots are used to create new logical units of data, each of which…

Read more

data

ZFS compression and Permabit Albireo COMPRESS compared

| By: (56)

In my recent article ZFS dedupe just doesn’t make any sense, I questioned why anyone would use ZFS deduplication. I’m not going to ask that question about ZFS compression, from what I’ve read it’s a well designed feature that for some use cases can actually speed up the performance of spinning disk. What’s missing from ZFS is frankly functional deduplication and that’s a shame because fast deduplication can actually speed…

Read more

speedo

VDO at One Million IOPS: Zero Hour for Tier Zero

| By: (56)

Yesterday Permabit announced that Albireo VDO has achieved 1 million sequential IOPS and 500 thousand random IOPS, which (to the best of my knowledge) are the highest performance numbers achieved by anyone, ever for a storage system with deduplication.  As a follow on, I wanted to provide some background on why we bothered to collect these numbers, how we got them, and what they mean to the industry as a…

Read more

ZFS Dedupe Just Doesn’t Make Any Sense

| By: (56)

Hey ZFS fans, let me know if I’ve got any of this wrong.  I mean no offense here, but not many people seem to be using ZFS deduplication in primary storage environments and if my math is right, it’s no wonder why! A couple years back, Constantin Gonzalez, then a consultant over at Oracle, wrote a blog titled “ZFS: To Dedupe or not to Dedupe”.  In it he goes through…

Read more

CATEGORIES: Deduplication

Save 10X network bandwidth with Permabit’s new REPLICA software

| By: (292)

  Albireo REPLICA is a great new product from Permabit that combines file-copying functionality and industry-leading deduplication in an innovative way.  REPLICA works in conjunction with Permabit’s VDO (Virtual Data Optimizer) extending VDOs capability into a richer data management product.  This post will describe REPLICA and share the results of some test provided by the Permabit Labs team that illustrates fantastic 10x advantage over other file transfer utilities.  In the same…

Read more