Workloads Performance & Optimizations; Target Workloads For Ultrastar Memory; Table 5-1 Workloads And Applications - Western Digital Ultrastar DC ME200 Product Manual

5.0 W
ORKLOADS
5.1

Target Workloads for Ultrastar Memory

The Ultrastar memory drive can take advantage of one or more of the following
workload characteristics to achieve DRAM-like performance, which can be associated
with the sample relevant applications:
Table 5-1.
Probability-based memory
access pattern
[pre-fetch]
Predictable workloads:
Analytics, DBMS, etc.
Memory access can be
predicted by:
Frequently similar access

pattern (sequential,
structured, etc.)
Application code analysis

Memory block associativity

Examples can be:
Databases tuples

In-memory columnar

database
Reading database indexes

Large in-core matrices

Block of memory stored from
a disk through DMA.
While relevant application
workloads can be row- or
column-store in-memory
databases used in analytics
workloads, such as SAP
®
HANA
MySQL
Workloads Performance & Optimizations
P
ERFORMANCE
Workloads and Applications
®
®
, Oracle , 12c, or
TM
& O
PTIMIZATIONS
High concurrency access
pattern [asynchronous
memory load]
Many processes, or highly
multi-threaded.
Concurrency can be due to:
Many threads

Throughput

(many independent jobs)
Multi-tenants

Relevant application
workloads can be:
Multi-tenant workloads

like container-based
virtual-shared web-
hosting server with
Docker , or
TM
Virtualization-based
partitioning for example
with KVM.
Multi-threaded key-value

cache such as
memcached.
Distributed/shared data

grids and frameworks
®
such as Apache Spark
®
Apache Ignite ,
Aerospike, or Redis .
TM
- 30 -
CPU intensive
[placement]
CAE, HPC
Those are workloads
which are heavy on
compute vs. memory
access. Relevant
applications and
workloads can be:
Multi-threaded linear

algebra workloads
with large matrices.
Parallel statistics

calculations on large
data.
,