Friday, March 1, 2024
HomeTechnologyPoINT provides on-site S3 object storage archives on disk and tape

PoINT provides on-site S3 object storage archives on disk and tape

Archival Gateway can cluster up to 256 tape readers, provides S3 object storage clustering, can store 50 billion objects per bucket, and has a throughput of 230GBps

go through

  • Yann Serra,

  • Anthony Adhead,
    Storage Editor

    issuing time:

    September 26, 2022 15:00

    In October, PoINT, Germany, will launch an object storage archive product based on disk and tape media.

    “Our idea is On-site delivery with AWS [Amazon Web Services] provides online services via S3 and S3 Glacier,” said PoINT CEO and co-founder Thomas Thalmann.

    Archive Gateway – Unified Object Storage uses high capacity, low cost disks to store cold data. That is, data that businesses will only rarely access.

    Like S3 Glacier, tape is also used. It is slower than disk but also cheaper.

    point that can recover files on disk in milliseconds, Files on tape are in minutes. To determine which files go to which media, PoINT’s software analyzes more than 50 parameters using predefined rules.

    For example, a user might want to One type of file is migrated from disk to tape, or another type of file is deleted from the archive after a set expiration date. Alternatively, they may want to copy the files to both media, but keep them on disk as long as the analysis work is performed.

    Compared to archiving services in public cloud, PoINT’s service guarantees that data doesn’t cross corporate firewalls — instead of having to constantly pay cloud providers to keep data on their infrastructure, businesses own disk and tape devices.

    230GBps tape only

  • Archive Gateway – Unified Object Storage is the first product of this type to write objects to the bucket directly to tape.

    Most other object storage products have the ability to archive content to tape The way to do this is to take a backup and copy it to the tape library in file format. This makes the process more verbose and means that the recovered data will not be in object format. To retrieve object metadata and provide S3 access, it must be restored to the S3 cluster. This means that several stages in the operation are usually dictated by emergency situations.

    ” With Archival Gateway, you can build An S3 cluster of eight tape libraries, each with eight readers working in parallel,” Thalmann said. “Just like we did with erasure coding at the tape level, we achieved parallel access. This scheme, even without a hard drive, achieves a throughput of 230GBps.”
    Archival Gateway – Unified Object Storage is the second no HDD, only tape product launched in Germany in 2018 Version.

    “For our German customers, we The announced novelties are the arrival of tiering, less latency on disk, and going from one can pass about 50 rules

  • For customers elsewhere, Archive Gateway is a brand new S3 storage product that can manage up to 50 billion objects per bucket due to the large number of cartridges.

    Archive Gateway includes gateway nodes that provide S3 object storage on the network Each of these nodes has access to up to eight read heads. The gateway is backed up by so-called database nodes, up to four per cluster. These index contents help guide the read heads and run the admin console for the product.

    Descendants of Philips and Sony

    Behind the news, PoINT has a stellar history in archiving. Its product, Archival Gateway, is a descendant of another archival product the company developed for Sony in 2016, Everspan Gateway. At the time, it wasn’t a tape reader cluster, but Consists of 64 Blu-ray Disc recorders with 18GBps performance and 181PB of data indexed.

    PoINT first flourished in the mid-90s with CD recorders, one of which – CDWrite – was a worldwide success on Windows. Before founding PoINT in 1994, Thalmann and his team had Worked at Philips, where they developed the basics of CD-R, the first CD-ROM that could be written.

    In addition to Archival Gateway, PoINT has sold another storage product since 2007: PoINT Storage Manager. This identifies production data on the network, and which is which are used and which are cold. According to this analysis, it moves data to different storage tiers and at the end of its life cycle, to disk or Tape archive.

    PoINT Storage Manager claims two unique properties. The first is that it leaves a symlink on the production array to the moved data. This means that users can still see files where they expect them to be, even if they’ve been layered elsewhere.

    “These are not simple symlinks in the Windows or Linux sense, ” Thalmann said. “These links use proprietary APIs from a number of NAS arrays including NetApp and Dell EMC.

    ” This system allows us, when the user wants to reopen a file, to gradually zoom in on the desired chunks. Our archives are immutable and resistant to ransomware. If the user modifies the content, we generate a new archive with all the same links. “

    The second feature is the format of the archive, which is a A disk image with a file system in common disk format.

    “For schematization, most archiving solutions will package your data into a zip file or equivalent,” he said. “The problem is that you only have the original file in it. If you package content into disk images, you can also archive their creation, modification, read dates, and their authorship and access rights with them. “

    Read more about disaster recovery

    • AWS Snowball Edge adds offline tape migration

      After: Adam Armstrong

  • Cloud Archive JohnnyYu

go through: Dave Laffer

  • AdamArmstrong JohnnyYu XenData to tape archive with S3 interface

    After: Johnny

  • JohnnyYu Evaluating cloud-based object storage for backup


    Brian Posey



    Please enter your comment!
    Please enter your name here


    Featured NEWS