Forrester Total Economic Impact Study: Migrating a Heterogeneous Database Environment to IBM DB2 for SAP and Non-SAP Environments

Contributed by Sid Misra

Forrester conducted a study to evaluate the potential financial impact of migrating to DB2. The study illustrates the financial impact of adopting IBM DB2 for both SAP and non-SAP applications based on aggregated findings of customers that migrated to DB2 from a variety of database platforms like Oracle database.

In conducting in-depth interviews with five global organizations using IBM DB2 both within SAP and non-SAP environments, Forrester found examples of how DB2 delivered key benefits both inside and outside of IT within three key benefit areas: reduced cost of administration, improved operational expenses, reduced storage cost.

Key Findings include:

•       ROI:  Based on data from five DB2 customers, Forrester found the ROI for a representative organization is 192% with a breakeven point (payback period) of 12 months.

•       Benefits: Incremental benefits included improvements in storage and administration efficiency, lower operational and capital costs and savings of reduced storage through compression.

•       20-30% reduction in impact of backup and restore events on IT and end user organizations

•       30% reduction in growth of staffing

•       15-30% reduction in problem resolution time

•       15% reduction in maintenance cost

•       Reduction in growth of 30% per year due to compression, with compression rates between 35% and 75%

You can download the entire report at https://www14.software.ibm.com/webapp/iwm/web/signup.do?source=sw-infomgt&S_PKG=forrester_db2_tei

Watson is ready to go to work inside your company

“This blog post also appears in smarterquestions.org”

IBM’s transformational technologies, like Watson, are ready to go to work in your company’s technology environment.

Watson technology is helping power the superior analytics capability behind IBM’s portfolio of Warehousing and Analytics solutions, specifically IBM InfoSphere Warehouse and IBM Smart Analytics System.

Let the power of Watson drive business decisions. IBM’s Glen Sheffield shares his views on how this exciting technology is impacting IBM’s warehousing and analytic solutions.

 

To learn more about IBM’s warehousing and analytic solutions:

IBM Smart Analytics Systems vs Oracle Exadata X2-2

An InComparison Paper by Bloor Research, by Philip Howard Philip Howard, IT analyst with Bloor Research in the U.K., provides a comprehensive comparison of IBM and Oracle systems for data warehousing.

IBM DB2 9.7 and Smart Analytics Systems

Architecture and Key Capabilities for Data Warehousing

In this detailed whitepaper, Winter Corporation concludes that “Overall, DB2 9.7 and IBM’s Smart Analytics Systems are among the most capable data warehouse products available on the market today.”

Simple, Flexible and Customizable!!!

Contributed by Jagadeesh Sampath

Last week I was shopping in an electronic shop with a friend, I was looking for a small travel iron box and my friend was thinking of picking up a steam iron. We enquired with the sales person and gave our requirements. He showed us a particular model and was trying to convince both of us to buy the same product. But we tried to explain him that we have different requirements and this one model won’t fit both of our requirements. Finally, we walked out of the shop as that particular model didn’t meet both of our requirements.

If that’s the case in buying small electronic products then what’s the case of executive’s making big purchasing decisions for large enterprises. Every business has different requirements, complexity and data volume; one solution can’t be addressing different business challenges in real world. When I specifically start talking about Data Warehouse, every business has different requirement depending on aforementioned factors.

There are Data Warehouse providers in the market offering solutions that are not for any specific requirement. Whatever is your requirement they tend to come up with the same solution. They say it can do Data Warehousing, Online Transaction Processing, Operational Data Store and what not. What happens is the system becomes very complex as it is designed to meet the requirements for everyone and thus won’t fully meet the requirements of anyone. When it is not fulfilling your requirement then what’s the use of it doing everything in the world.

So, here comes the concept of “Workload Optimized Systems” that are purpose built systems to cater for different requirements of businesses. IBM has clearly understood this and offer different Data Warehouse solutions to address different workloads.

• “Smart Analytics Systems” combine necessary hardware, software, analytics and services; it can grow as and when your data requirement grows
• “InfoSphere Warehouse” can be customized according to your needs, you can add Cognos and SPSS to fulfill your analytical requirements and add necessary hardware as well
• In addition to this IBM also wanted to offer Data Warehouse appliances to cater the needs of customers who had a plug-and-play kind of requirement. So, recently acquired Netezza, a leader in Data Warehouse appliances

Customers can now choose from a wide range of Data Warehouse solutions (Smart Analytics System, InfoSphere Warehouse and IBM Netezza TwinFin^TM ) according to their business requirement. IBM offers right mix of simplicity and flexibility to the Data Warehouse customers.

IBM Data Warehouse & Analytical Solutions – “Simple, Flexible and Customizable”

For more information please visit: http://www-01.ibm.com/software/data/infosphere/data-warehousing/

Watson making Information Management (even more) cool

“This blog post also appears in the IBM Research News blog”

Contributed By Bernie Spang

Like many of my colleagues in the IT business, I am often disappointed by the lack of interest I am able to generate among my family and friends when discussing my work. But thanks to Watson, the computing system that can play quiz show Jeopardy! at a champion level, I have experienced a few precious moments where both my kids and parents showed interest in my work

My 15 year old son and 76 year old father both had the same reaction after watching the Watson-Jeopardy! Challenge commercial during the NFL playoffs: “that’s cool, but why did IBM build a computer to play games?”

Surprisingly, they both gave me five minutes of their attention – just long enough to sneak in an explanation of how Watson is connected to the IBM Information Management software portfolio.

What is Information Management?

Watson incorporates open technology such as UIMA (Unstructured Information Management Architecture), Eclipse, and Apache Hadoop. The first two of which IBM contributed to the open source community. To build on the medical reference IBM Research Senior Vice President Dr. John Kelly made at the Watson-Jeopardy! press conference in January, a healthcare provider can use this software to analyze patient and treatment information – including doctors’ notes and clinical reports – to pinpoint illness trends and successful treatments.

These capabilities Watson uses Apache Hadoop to analyze massive amounts of information is also used in IBM’s InfoSphere BigInsights software. And IBM InfoSphere Warehouse uses the UIMA technology for text analytics – just as Watson does.

A new addition to the InfoSphere portfolio, called Streams, analyzes information flowing through systems that may never be stored. Streams can analyze thousands of pieces of vital sign telemetry per second to help save the lives of premature babies (another compelling example that kept my son and father intrigued). While Watson does not use a form of Streams, the two have shared heritage as IBM Research projects.

If you are anything like my dad and son, you are about at your limit of examples to absorb. But hopefully you, too, already get the point. While Watson is an amazing feat of Question Answer technology, my son and dad think the future possibilities for Watson are pretty cool, too.

Oracle Confirms that IBM is 3x Faster

Contributed by Sid Misra

Oracle/Sun again in a desperate attempt to react to its losing share in the server market, announced a new record TPC-C benchmark result for a cluster of SPARC systems on 2nd December 2010. While on the surface the result appears bigger than the recent IBM TPC-C result on the tpmC metric but just looking at some basic details of the result exposes how inefficient Sun/Oracle systems are.  Read on for more details.

The Oracle benchmark result uses 27 64-core Sun SPARC servers to process more than 30M tpmC*. In contrast, IBM’s most recent clustered TPC-C result uses 3 64-core IBM Power Systems servers to process more than 10M tpmC**.
IBM’s result is better overall for the following reasons:

• Better throughput per core – The DB2 on Power Systems result was run on three 64-core IBM Power 780 servers while Oracle’s benchmark result uses 27 64-core Sun SPARC servers. With 9 times more processor cores, Oracle only beat the throughput of the DB2 on Power Systems result by three times. In other words IBM DB2 on Power Systems delivers over 3x more throughout per core than Oracle Database on Sun SPARC.
• Better Performance per Server – Oracle uses 27 servers to achieve this result.  The IBM TPC-C result on the Power 780 requires floor space, energy, and systems management for only 3 servers. This is a lot less floor space, energy, and systems management.
• More Cost efficient – Since customers pay for database software licenses by the core, you may lower software license and maintenance costs with a system that has less cores for comparable performance.

Further, these TPC benchmark results are consistent with recent findings in the two-tier SAP Sales and Distribution (SD) benchmark, where IBM’s leading benchmark result supports 126,063 SAP SD benchmark users***, which is more than three times the number of users of Oracle’s leading result****. Again, an IBM benchmark system has been independently verified as being three times more efficient than a comparable Oracle benchmark system.

Well, Mr. Ellison, given all the misleading information and questionable results from Oracle, you may need to swap the Cheetah and the Stallion on your slides.

Results on Transaction Processing Performance Council Web site at http://www.tpc.org. Results as of 12/02/10.

* Oracle SPARC SuperCluster with T3-4 Servers (27 x 64 core) (108 chips, 1728 cores, 13824 threads); 30,249,688 tpmC; $1.01/tpmC; available 6/1/11.
** IBM Power 780 cluster (3 x 64 core) (24 chips, 192 cores, 768 threads); 10,366,254 tpmC; $1.38/tpmC; available 10/13/10
*** IBM Power 795, 32 Processors / 256 Cores / 1024 Threads; 126063 users; 0.98 average dialog response time; 688630 SAPS; certification number 2010046; 11/15/2010.
**** SPARC Enterprise Server M9000, 64 Processors / 256 Cores / 512 Threads; 39100 users; 1.94 average dialog response time; 196570 SAPS; certification number 2008042; 7/14/2008.

 

When will the database world wake up?

Contributor: Jagadeesh Sampath

As the IT budgets are shrinking and the expectation for high performance is increasing considerably, customers are looking for databases that offer better performance at an optimum price. So, what should you do? Start looking at the different database options available in the market before proceeding to purchase a new database or upgrading an existing one. The perception in the market is that not many database options available and I think this need to be changed. Moreover, there are lots of myths in the market that alternate database options are expensive, unreliable, inefficient, complex etc. All these are really true?

Let me take you through a detailed analysis on IBM DB2 –The tried and tested database from IBM and try to address some of the question on your mind.

IBM DB2 has a long history, the Database Management System (DBMS) was first named DB2 in 1983 and it has come a long way after that. In, 2008 IBM DB2 completed its 25^th year anniversary and it is still going strong. IBM DB2 is a full-function, on-disk, relational database that runs on distributed platforms (Linux, UNIX, and Windows) as well as on z/OS (IBM’s 64-bit mainframe operating system) and i5/OS. It is optimized to handle both online transaction processing (OLTP) and analytical workloads with lower administration, storage, and development costs. IBM DB2 powers IBM InfoSphere Warehouse and Smart Analytics System which offers data warehouse capabilities.

Let me start with the two key factors: performance and costs. IBM DB2 has been consistently showcasing better performance leadership compared to other databases in market. I would like to put below some of the key performance benchmark (Transaction Processing Performance Council (TPC) – Industry standard bench mark) results of IBM DB2,

• #1 in the 10 TB TPC-H Performance (a data warehouse workload)
• #1 in the TPC-C Performance (an OLTP workload)
• #1 in SAP SD 3-tier Standard Application Performance

I would also like to show the number of days of leadership for IBM DB2 against Oracle Database since Jan 1, 2003:

You might ask me: Ok, the performance is better than the competition but what about the cost. Does better performance comes with a high price tag attached with it? No, absolutely not. IBM DB2 is more cost efficient than competing products especially Oracle Database. IBM offers a capacity on demand pricing/licensing model that eliminates over-provisioning by allowing you to pay for peak workloads only when needed. IBM DB2’s better performance also helps in cutting costs, as better performance will delay server purchases, since you get more done with your current infrastructure. In contrast, Oracle pushes you to buy more than what you need through the suite model and doesn’t offer discounted licensing for non-production environments. Moreover, Oracle has raised its database prices four times since June 2008.

You can get better price-performance with IBM DB2 compared to other databases in the market. I know as a buyer these are not your only consideration. So, I will also put across some of the key benefits of IBM DB2 and try to compare it with competition especially, Oracle database to show how IBM DB2 is better than competition.

IBM DB2 offers Deep Compression, a compression technique that looks for repeating patterns across the entire table and compresses it. The benefit of deep compression is that a table can shrink by 60% to 80% of its original size. Therefore, you will see significant I/O bandwidth savings and thus, better performance. On the other hand, Oracle also offers compression capability but it compresses out common values at the page level not on the table level. Thus, the compression achieved on Oracle database is far less compared to IBM DB2 Deep Compression.

Any enterprise will look for high availability as it is a critical feature in a database. Knowing this IBM offers High Availability and Disaster Recovery (HADR) in IBM DB2. HADR is a log record replication feature that provides high availability for server, software or storage failure, or disaster recovery for site failures. HADR is designed for quick failover, easy setup, and manageability. While, Oracle’s own users do not rely on Oracle RAC for availability, they build a redundant and idle site using DataGuard. You have to pay for the “idle” – CPUs, servers, and storage, but won’t benefit from that investment.

Better price-performance, better database compression, HADR: are these the only value proportion to buy or move to IBM DB2? No, IBM DB2 offers more values to the customers.

IBM DB2 offers new optional feature called DB2 pureScale. This feature reduces the risk and cost of business growth by providing unlimited capacity, continuous availability, and application transparency.   DB2 pureScale allows you to have multiple database servers in a system that all share a common set of disks. So, now you can buy only what you need, add capacity as your needs grow. Tests show that the scale-out efficiency was above 95% for up to 65 members in the cluster and at 84% for 128 members.

Next is the IBM DB2’s pureXML technology, a superior technology compared to the Oracle Database XML storage capability.  IBM DB2 store XML data natively in the same database as your relational data and do not force you to have a schema per XML column. Also, allow you to have multiple schema for the same XML column. Moreover, IBM is the only database vendor with detailed published XML performance results.

I was keeping the best for the last and I need to talk about this now: The IBM and SAP partnership. Strong partnership exists between IBM and SAP, joint SAP and IBM teams work on all levels of the product development. DB2 is certified by SAP within 4-8 weeks of its general availability (GA) date while Oracle does not have its new database releases certified by SAP until 1.5 to 2+ years after their GA date. All new DB2 features are designed to be 100% compatible and exploited by SAP directly. A joint technology roadmap exists beyond 2010.

I think it’s a right alarm at the right moment. It’s high time you look at the database options available in the market before making a purchasing decision. This might prove critical to your business in the long term. So, “RETHINKDATABASE” strategy is the way ahead.

http://www-01.ibm.com/software/data/db2/

If you think it isn’t broken, maybe you are not setting the bar high enough

While the old adage – if it isn’t broken, don’t fix it – is often the right advice to follow; it can also be the most costly answer.  Those who fail to take advantage of “upgrading” old systems to more efficient, reliable and higher performance solutions can be missing an opportunity to save operational costs and eliminate time consuming maintenance work.  For those of us who have faced the choice of continuing to pay growing maintenance and fuel costs for an aging automobile that “still works”, or upgrading to a new one, we know the overall cost advantages are typically what justify the decision to upgrade.  The better performance and new convenience features are benefits we then enjoy in addition to saving both time and money.

Many clients that I speak with are facing a similar decision when it comes to their database software.    But many more are not even aware they are in this situation; or that there are multiple upgrade alternatives that could help their business save time and money.  Most still believe that updating versions of existing database software is the only cost effective, low risk option available to them.  But there are a growing number who have reduced operating costs and improved application performance and reliability by upgrading from Oracle Database to IBM DB2 software.

In 2009 IBM delivered DB2 9.7 with a new capability referred to as SQL Skin for Oracle Database applications.  This is not a clunky conversion or emulation layer; DB2 now implements the most commonly used SQL syntax, PL/SQL procedure language and other non-standard characteristics of Oracle Database.

DB2 added a SQL Skin for Sybase ASE applications in 2010, providing a low cost, low risk path to DB2 for those looking to upgrade from Sybase ASE.

So now we are seeing a growing number of new clients who are cutting costs and improving performance by moving to DB2.  The big reason, in addition to the consistently proven better performance and lower cost, is that switching database software no longer means the cost and risk it used to have.  Many of those who needed to augment the skills of their database professionals also took advantage of IBM’s no charge training:
http://www.ibm.com/developerworks/wikis/display/im/DB2+9.7+Bootcamp+and+Oracle+to+DB2+9.7+Migration+Clinic

Analysts who have taken note of the importance of this innovation include Forrester’s Noel Yuhanna,
http://blogs.forrester.com/noel_yuhanna/10-06-30-database_migrations_are_finally_becoming_simpler

and Gartner’s Donald Fienberg.
http://www.gartner.com/DisplayDocument?doc_cd=167786

Another example I could offer for not setting the “it isn’t broken” bar high enough, would be accepting the high cost use of Microsoft SQL Server vs upgrading to IBM Informix software.  How could SQL Server be higher cost, you ask?  When the performance and reliability upgrade means you can eliminate dozens and dozens of Windows servers to save cost of space, power and administration with much fewer Linux servers.  Not to mention the number of DBAs you could reassign to more productive work as you won’t need them to spend anywhere near as much time on managing Informix databases.

And speaking of Informix, another surprising case I often see is a situation where someone thinks something is broken and proposes “fixing it” by changing to a solution that costs more and delivers lower reliability and performance.  I think I will save thoughts on the topic of “If you think it is broken – why are you making it worse?” for my next entry.

It’s time to reTHINK database.

Welcome to the reTHINKdatabase blog.

The goal of this blog is to provide a unique perspective on IBM strategy and marketing in the business of data. Why, you ask? Why now?

You may recall that earlier this summer, CEO Palmisano laid out plans for 2015 that included a $20 EPS target, essentially doubling the company’s profit in five years. That means only one thing – IBM is going to get very aggressive in its growth strategy that includes a greater focus on high value segments, growth markets and acquisitions ($20 Billion to be invested in the next five years)

Among those high value segments, Palmisano identified software, analytics and Software-as-a-Service as sweet spots. The contributing authors of this blog are marketing and strategy leaders of exactly that chunk of the IBM business, which gives us front row seats to the best action. Interestingly, this space has heated up in the last eighteen months.

The timing of this blog is critical. IBM has a long history of delivering scalable, adaptive, integrated software-hardware systems to solve critical business problems. More recently, IBM has further strengthened its commitment to integrated, optimized systems (or Smarter Systems) by announcing an organizational realignment that essentially merged the hardware and software businesses into one. Through this blog we plan to bring you an insider’s view of IBM’s business engine as it relates to the Smarter Systems business.

Thanks for stopping by and we hope to see you back soon. Please post comments on topics you’d like to see us blog about. And yes, we’re looking for guest bloggers; if you’re interested please drop us a note.

ps: Check out the reTHINKmigration video – you’ll see we also know how to have a bit of fun.