Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
29 Cards in this Set
- Front
- Back
What are the two components of a block-based storage system?
|
Controller (FE ports, Cache and BE ports) and Storage |
|
Read hit
|
Data is found in cache following a read request
|
|
Read miss
|
Data is not found in cache following a read request |
|
Prefetch |
A read-ahead algorithm that calls sequential data into cache to improve read time |
|
Write hit
|
Following a write request data is dumped in cache and an acknowledgment sent back to the compute system. Data is written into backend storage later on. (Also known as write-back cache) |
|
Write miss
|
Following a write request data is immediately written to backend storage, no time is spent waiting on cache. (Also known as write-through cache)
|
|
LRU
|
Least recently used. Is an algorithm used to free up cache that has not been accessed recently |
|
MRU |
|
|
Idle Flushing
|
Dumping cache into storage, continuously at a modest rate |
|
High Watermark Flushing (HWM)
|
When cache utilization hits a threshold causing it to speed up the flushing process |
|
Dirty Page
|
Data written into cache but not yet written to backend storage
|
|
Forced Flushing
|
When cache utilization reaches 100% the system will flush the cache on priority by allocating more resources |
|
Cache Mirroring
|
Each write to cache is held in two different memory locations on two independent memory locations |
|
Cache Vaulting
|
When power failure occurs the controller and a few discs remain powered on by battery until all Write Pending cells are flushed into storage |
|
Storage Provisioning
|
The process of assigning storage resources to compute system based on capacity, availability, and performance requirements |
|
LUN
|
Logical Unit Number. A partitioned chunk of storage from the RAID set |
|
MetaLUN
|
A method to expand LUNs that require additional capacity or performance |
|
Concatenated metaLUN
|
Adding additional capacity to the base LUN. Once the base LUN is full new data will be written to the extended capacity. LUNs can be variable size and RAID levels |
|
Striped metaLUN
|
Striped expansion involves restriping the base LUNs data across the newly added LUN. This can take time to configure but will improve performance due to increased number of striped LUNs |
|
LUN Masking
|
A process that provides data access control by defining which LUNs a compute system can access |
|
Storage Tiering
|
A technique of establishing a hierarchy of storage types (tiers) and identifying the candidate data to relocate to the appropriate storage type to meet service level requirements at a minimal cost |
|
Sub-LUN Tiering
|
A LUN is broken down into smaller parts that are separately tiered based on performance |
|
Cache Tiering
|
SSDs are used to create a large capacity secondary chance to enable performance tiering between DRAM and SSDs
|
|
Server Flash-caching
|
a PCI Express-based flash card is installed directly on the compute system which reduces latency and accelerates throughput
|
|
XtremIO
|
-All Flash -Block-based -High I/O performance -Node based architecture for scale-out -Operating system: XIOS |
|
VNX-F
|
Block-based SSD only storage
|
|
FAST VP
|
-Moves data that is not frequently used to lower tiers and vice versa |
|
XtremSF
|
-PCIe flash card deployed on the compute system -Used in conjunction with XtremCache |
|
XtremCache |
-Protects data with write-through cache to the networked storage |