SDI_halukUlubay_4by3_28NOV2015-V2

12
Software-Defined Storage Makes Economic Sense Haluk Ulubay Sr. Director, Marketing, DataCore Software Corp. Santa Clara, CA USA December 2015 1

Transcript of SDI_halukUlubay_4by3_28NOV2015-V2

Software-Defined Storage Makes

Economic Sense

Haluk Ulubay

Sr. Director, Marketing, DataCore Software Corp.

Santa Clara, CA USA

December 2015

1

Agenda

What is Software-Defined Storage (SDS)

Distinguishing Characteristics

Risks

Top 3 SDS Reasons

Check list

Santa Clara, CA USA

December 2015

2

SDS Definition – A Sampling

• Wikipedia: “an evolving concept for computer data storage software to

manage policy-based provisioning and management of data

storage independent of the underlying hardware.” - Wikipedia

• SearchSDN: “an approach to data storage in which the programming

that controls storage-related tasks is decoupled from the physical

storage hardware.”

• IDC: “platforms that deliver the full suite of storage services via a

software stack that uses (but is not dependent on) commodity

hardware built with off-the-shelf components.”

• Webopedia: “Storage infrastructure that is managed and

automated by intelligent software as opposed to by the storage

hardware itself. In this way, the pooled storage infrastructure

resources in a software-defined storage environment can be

automatically and efficiently allocated to match the application needs

of an enterprise.”

Santa Clara, CA USA

December 2015

3

Distinguishing Characteristics

Separates advances in software from advances in

hardware (i.e., abstraction)

Stand-alone, intelligent software that runs on multiple

hardware instances

Does not require proprietary hardware

Provides a common set of storage services for all

types of storage devices / systems underneath

Pools all storage capacity and provides centralized

management

Santa Clara, CA USA

December 2015

4

Risks for IT Managers

Various & sometimes narrow implementations of

SDS that:

• Only applies to a use case (e.g., Virtual SAN, hyper-

converged)

• Are controllers that require specific hardware

• Is confined to limited types of storage

• Is tied to a specific hypervisor

• …

Santa Clara, CA USA

December 2015

5

Top 3 SDS Drivers

Santa Clara, CA USA

December 2015

6

In a survey that was published in April 2015 (*), the following was reported:

“What are the business drivers for implementing Software-Defined Storage?”

(*) Survey conducted by DataCore Software Corp. n = 948

40% 42% 44% 46% 48% 50% 52%

Extend the life of existing storageassets

Avoid hardware lock-in

Simplify management by automatingstorage operations

52%

49%

45%

Financial services

12%

Healthcare 12%

Government 13%

Manufacturing

11%

Education 9%

IT services 17%

Other 26%

INDUSTRY

< $10M 28%

$10M - $100M

34%

$100M - $1B 19%

> $1B 19%

ANNUAL REVENUE

Hardware Abstraction

Maybe the most obvious benefit: “Choice to choose”

Ability to utilize different vendors’ storage hardware

(side by side or interchangeably), including the server

side, current and future

• As long as the SLA is met for the particular application that is

asking for the “service”

Brings future investment protection & lower CAPEX

Features such as auto-tiering make this easy as well

SDS software running on any x86 platform is icing on

the cake

Santa Clara, CA USA

December 2015

7

Costs: CAPEX vs OPEX

Santa Clara, CA USA

December 2015

8

Considering a heterogeneous storage

environment:

Hardware / licenses / maintenance

contracts is CAPEX and can be

amortized over useful life.

Significant amount of OPEX / TCO

come from administrative work

Best to select an SDS solution that

will minimize operational costs • Data migrations

• Hardware maintenance and refresh

• Planned and unplanned downtime

• Capacity expansion

Source: Jon Toigo, Optimizing the economics of storage: It’s all about

the Benjamins. 2015.

Management

“Management of storage infrastructure” is

also referred to by most definitions.

All storage pulled into a virtual pool,

eliminating wasted capacity (i.e., no

siloes).

• Thin-provisioned

Automating and collapsing tedious &

manual tasks

Single pane of glass to manage all

storage in the pool provides for:

• Simplified configuration and management,

including SANs and cloud storage

• Lower OPEX (less training, personnel,

support of different management

systems)

• Multiple sites Santa Clara, CA USA

December 2015

9

What Else Could Come w/ SDS?

Data Protection / High availability storage

• Synch mirroring, snapshots / backups, auto fail-

over & fail-back

• Live maintenance (upgrades, migrations)

DR

• Asynchronous replication over long distances

Application performance

• Caching

• Auto-tiering

• Parallelism where applicable

Santa Clara, CA USA

December 2015

10

Checklist

• Do I need to buy proprietary hardware to run the SDS software?

• Can the SDS solution pool and manage all of the heterogeneous

hardware systems that I have now (and the ones I am planning

to add in the next 12 months?)

• Does it provide me with a single management interface with

which I can provision (manual and auto) and manage all storage

systems that I have in multiple sites?

• As a minimum feature set, does it have virtual pooling, auto-

tiering, no-touch fail-over / fail-back (make your list)

• Can I do in-service (hitless) planned maintenance, s/w or h/w

upgrades, migrations? How many nodes are required to do that?

• How does it handle i/o bottlenecks (that will slow down business

critical applications?)

• What benchmarks will help me choose between alternatives?

Santa Clara, CA USA

December 2015

11

THANK YOU!

Haluk Ulubay

[email protected]

www.datacore.com

Santa Clara, CA USA

December 2015

12