![]() ![]() the controller says hey let’s make a copy of that and put it in the SSD cache. Say this same piece of data is repeatedly read/requested, over and over. If not, then it checks the SSD cache to see if it’s there, and if it isn’t goes to the HDD for the read and sends it up. A request for a certain block of data comes in and the controller checks it’s onboard DDR cache to see if it is there. That describes the write side of the house.įor the read, it’s the same thing but slightly different. As the writes to the HDD complete, the cache likely flushes itself. ![]() The whole time these two caches are getting filled up, the HDD array is writing as quickly as possible to the HDDs, playing catch-up to the solid state based caches (the onboard DDR and SSD caches). If the onboard controller cache fills up because the file is larger than the cache, it will direct the writes to the SSD based cache. So once you have it setup, writes intended to go to the HDD array will first be written to the DDR cache on the controller, which is trying to quickly write it to the HDD. ![]() The CacheCade feature allows for a regular SSD connected to the raid card normally to be identified as the read/write cache for slower, but likely higher capacity and cheaper per Terabyte, HDD based arrays. ![]() So basically it’s a feature that gets enabled through the card when this little key is attached to the raid controller. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |