Posts Tagged ‘integrated copy data management’

You Don’t Have to Divide – Just Rule the Database Workloads

Bala Chandrasekaran

Principal Product Marketing Manager, Dell EMC
Bala Chandrasekaran is a Principal Product Marketing Manager at Dell EMC and delivers marketing efforts for All Flash primary storage. He brings to Dell EMC his 12 years of IT industry experience across various customer facing roles including pre-sales, consulting and technical services. He is a certified professional in IBM DB2 and Oracle Databases. He has an undergraduate engineering degree from Anna University, India and MBA from Foster School of Business of University of Washington.

Latest posts by Bala Chandrasekaran (see all)

In the digital economy, there is immense value in drawing real-time business insights from organizations’ own transactional systems. However, it is not uncommon for a business to wait several hours before actual sales report or the inventory statuses are made available.

Typically, enterprise IT divides business processes into online transaction processing (OLTP), batch and analytic/reporting workloads. In order to meet the performance and scalability needs of these workloads, IT teams built systems that are optimized to execute one or the other. This complex ‘divide and conquer’ approach for database workloads leads to both business and IT complexities. For example, complex queries involved in reporting workloads would require indexes to produce reports quickly but having too many indexes on a transactional table would cost OLTP performance.

One of the top reasons our customers choose us as they modernize their data centers is the value our comprehensive portfolio offers. With our All Flash storage, our customers not only reduce the batch reporting duration but also enable new business process workflows along with integrated Copy Data Management (iCDM) solution.

The scale-out architecture of our storage systems enables them to consolidate multiple database workloads and comfortably deliver consistent performance to the applications. With virtual copies technologies, such as XtremIO Virtual Copies (XVC), database copies for workloads such as reporting and testing could be provisioned without traditional time overhead.


As shown above, with the full-featured AppSync Starter bundle, our customers are able to automate the entire workflow of making and provisioning application-consistent database copies in 3 simple steps:

  1. Centrally define source hosts and let AppSync automatically discover the databases.
  2. Customize the database cloning process to prepare the database for the target workload:
    1. AppSync allows administrators to add or modify database initialization parameters ideal for the target workload. Do you need to use a larger value for DB_FILE_MULTIBLOCK_READ_COUNT to improve IOs on table scan? No problem!
    2. AppSync can execute user scripts before making a copy or after mounting a database. Do you need to create that index required for your complex queries involved in reporting? No problem!
  3. With the ability to define service plans, administrators can automate the creation and deletion of virtual copies.

As decision-making windows shrink for most digital businesses, they require a different approach to retrieve information out of their own transactional systems. Storage plays a vital role in enabling data agility and with the right solution, IT teams can enable information flow that modern-day businesses require.

Drowning in Copy Data? It’s Time to Bring Order to the Copy Data Chaos

Caitlin Gordon

Director, Product Marketing, Storage Division
Caitlin has spent the past eleven years focused on all things storage and data protection, with a focus on enterprise storage. When she is not blogging, Caitlin spends her energy on her other passions as a tech nerd, sports nut and mama bear. With her “free time” (aka when she’s not chasing down one of her two young kids), she’s researching and discussing the latest tech rumors or celebrating/lamenting the latest Boston sports heartbreak/victory. You can also find Caitlin on Twitter (@caitsgordon) and YouTube.

The Copy Data Problem
Are you tired of hearing about data growth?   Are you even more tired of coping with it?  The dirty little secret of this unabated/exponential/non-stop data growth is that it’s being driven by copy data.  Copies could be anything from a snapshot, backup, replica or a copy made for test/dev, analytics or operations functions.  According to IDC, up to 60% of storage capacity is dedicated to copy data today and it will cost business over $50B in 2018.

In fact, a recent study shows that 82% of businesses have at least 10 copies of any one production instance of data.   So it’s no surprise that the vast majority of organizations expect to spend more on storage this year than last.   What you probably don’t know is that the #1 reason that organizations expect to spend less this year is the use of effective copy data management.   Unfortunately, that’s a small minority (20%) of businesses today, which is why we still can’t stop talking about or dealing with data growth.  So what can you do about it?


The Answer
The copy data problem is too broad and complex to solve with a one size fits all approach.  It requires a strategic approach that will ultimately drive down costs and curb storage growth.

First, you need to optimize your infrastructure efficiency to ensure you’re getting the most out of your storage and data protection assets.  That requires storing the right number of copies on the right storage devices and ensuring that data is both protected and accessible.  You will need to have a combination of all flash primary storage with incredibly efficient copy creation as well as protection storage, which minimizes storage requirements while ensuring recoverability.

Next, you need to balance the autonomy data owners crave with control of IT resources to streamline operations.   This is especially critical because a major driver of copy sprawl is the move to self-service copy creation with data owners (database admins, storage admins, VM admins, etc.) all creating copies from their native UIs.  However, without global oversight, this self-service model can massively proliferate copy sprawl.

Finally, you always need to meet service levels across the enterprise to minimize risk,– including data loss and downtime.  In fact, IDC estimates that downtime costs customers a million dollars per hour and our Global Data Protection Index found this adds up to nearly a one trillion dollar loss globally.  And this is expense that is totally avoidable by putting the right solution in place.


The EMC Copy Data Management Family
With the introduction of EMC Enterprise Copy Data Management today, EMC now offers a family of copy data management (CDM) solutions designed to complement our extremely efficient storage and protection technologies.   EMC’s CDM solutions are designed to comprehensively address the copy data problem and include:

Don’t drown in a sea of copy data. Let EMC throw you a life preserver with EMC CDM solutions.





Connect with us on Twitter

Click here for the Cloud Chats blog