Hello, my name is Tudor, technical marketing engineer in the storage grid team and today I will talk to you about why storage grid is the perfect complement for onremise on clusters. Cloud tiering powered by fabric poolool enables AFF and fast high performance storage systems to tier infrequently access data to storage grid and other S3based targets. The tearing engine works the same way for NAS L and snapshot data because the tiering takes place at the ONAP file system level which may be familiar to you under the waffle moniker. For an active file system, tiering is like a trust fault. The secondary tier must be there to catch it. And there is nothing scarier from a file system integrity point of view than being asked for some data and coming up empty. In other words, while tier data is by definition the less accessed data, it is still critical to the well-being of the active file system. If the secondary tier is missing, you have a sparse file system. Operations like directory listings and setting permissions on a folder that contains tiered and untiered data will hang when encountering a sparse file system. Please note how these metadata operations do not actually read the call data and that data would not have been recalled even if it was available. Now let's talk about ontap tiering and how it's fundamentally different in other ting solutions. One of the main advantages of onap tiering is that the file metadata always remains on the primary tier. Metadata only operations will continue to complete as if all the data is still on tier one. What that means for you is that operation like directory listings, finding directory sizes and setting permissions will not take longer just because data is now tiered off. And there are other architectural benefits as well. For example, the tiering engine evaluates waffle blocks after the storage efficiency engine has had a chance to ddup and compress data, which means that less space is consumed on the secondary tier. Defining what call data means is up to you. policies for tearing out and recalling data can be configured to target snapshot blocks only. Snapshot and active file system blocks or entire data sets. Additionally, ONAP is opportunistic when it comes to retrieving call data. Let's assume that the waffle block was copied to a secondary tier, but the original waffle block has not yet been overwritten by ONAP. A read request will ser the data from the primary tier, something that's only possible due to the deep integration between waffle and the secondary storage tier. Very smart. Finally, on tap tiering can be configured with two separate targets potentially in different sites. You can add or remove a secondary destination at any time and fabric pool will always keep them in sync. Storage grid is an ideal target for onapp tearing. An on-remise steering target provides the fastest and most robust tiering option. Multiple onap clusters, even geod dispersed, can tear to a single logical storage grid namespace because a storage grid instance can span multiple sites. Finally, when teing the storage grid, there is no need for a non-defab report or cloud ting license. Backup data is very important as well though in a different way than the immediacy of tier data. If you are missing tier data, you'll find out very quickly. If you are missing backup data, you will find out the worst possible moment. Having a reliable backup infrastructure is like having a parachute. You don't always need it, but when you do, you'll be glad to have it. Cloud backup gives ONAP administrators the best method of protecting ONAP unstructured data databases and virtual machines and it works at a block level very similar to tiering. As a matter of fact, the tracking of waffle block changes is a fundamental onep capability and a number of longestablished capabilities are based on it. Snapshot protection, snap mirror replication and caching are built on top of this foundation. Just like with tearing, all the onab storage efficiencies are preserved, which reduces the amount of storage used in storage grid. By using block level tracking and replication, cloud backup creates incredibly fast incremental forever backups. Cloud backup requires no media gateway and it gives users a fully automated backup plan that never impacts production and minimizes bandwidth use. Backup also can be made immutable for added storage security. Storage grid is an ideal target for onap backup data. It provides an air gap infrastructure with different authentication and management mechanisms and a single name space on top of geodispersed infrastructure. With this characteristics, it is fully compliant with any 3 to2:1 backup mandates. One single storage grid instance can replace multiple tearing targets and potentially multiple tearing solutions with a single system. Even if your onep clusters are geo dispersed, a single storage grid name space can span multiple data centers. Each storage grid site can scale differently, yet they all constitute a single entity managed centrally. One single storage grid instance can also replace all your tape infrastructure. Faster, more resilient, highly scalable, easy to upgrade from both the hardware and the software point of view. No more keeping all tape drives because you have tapes that still have to be read. No more the gaussing of tapes to erase sensitive data. No more manual operations to turn a tape into a read only device. Last but not least, a combined tier one, tier two, and archive solution from a single vendor has significant benefits as well. Better technical interperability, assured compatibility testing, and streamline technical support. Finally, when discussing licensing terms, renewals, and technical refreshes, working with a single vendor and cotterm licenses makes life just a little bit easier. Tier data and backup data are similar in that they do not belong on a primary storage system. At scale, the cost of storing cold and backup data on tier one storage far outweighs the savings of having a homogeneous storage infrastructure. Tearing and backup data have another common characteristic. While the performance of the data matters less, the data still has to be there. Alternatively, not having access to tiered or backup data are what we call ré generating events. So, how do you prevent that? With an infrastructure storage grid, which provides up to 15 lines of availability. Finally, as the saying goes, quantity is a quality in itself. Depending on how you define your call data, tier data could be up to 80% of your data set. As far as the backup data goes, even with incremental forever policies, you're likely generating large amounts of backup data. Storage grid is highly scalable and very cost- effective, especially when combining tearing a backup in a single grid. Titting data to storage grid does not require any ontap or cloud ting license. When using storage grid as the backup target, cloud backup licenses are a smashing deal. storage grid. There are no costs associated with data ingress or egress and the performance is significantly better than public cloud. Our takeaways today are simple. A single storage grid name space can be the target for multiple geodispers on tap systems. Both tiered and backup data can happily coexist in a single grid. Storage grid ILM policies can be used to maximize response times for tiered data and storage efficiencies for backup data. Storage Grid provides ample methods to secure and retain data long term which makes storage grid the perfect ONAP companion. Thank you for your attention and check out our community site for more of my blogs where you will find additional details on the synergies between ONAP and storage grid.