Ever since SAP first introduced its new in-memory appliance at the company’s annual SapphireNow conference last year, it has sought to convince users of the technology’s ability to simplify IT landscapes and process massive amounts of data in seconds.
SAP high-performance analytic appliance (HANA) is a data warehouse appliance for processing high volumes of operational and transactional data in real time. HANA uses in-memory analytics, an approach that queries data stored in random-access memory (RAM) instead of on hard disk or flash storage in order to reduce the latency time for bringing data to a processor.
While some customers have balked at being told they need to once again spend money for faster analytics when they already have spent money on things like SAP Business Warehouse Accelerator (BWA), others, like Mridula Sharma have listened.
In-memory analytics could help Sharma’s employer, a public utility in Snohomish, Wash., process reams of data on subscribers’ usage patterns so that they better plan for peak demand periods, she said. At the same time, it could help them greatly simplify their operations, she said.
“The business warehouse is based on traditional technology, where we take the data from the main core systems and take it into the business warehouse” she said, after which the date is then kept in data containers called InfoCubes.
“You can actually bypass all of those steps [since] in-memory computing takes data right from your legacy systems and puts the summarized strategic reports right out there. The development time is reduced by quite a bit. That is definitely interesting,” Sharma said.
“Not having to store five years of duplicated data in the cubes,” she added, “means we can save quite a bit on space.”
While a lot of attention has been paid to SAP HANA and high-speed analytics, it’s worth remembering that in-memory technology isn’t new, according to Rita Sallam, an analyst with Gartner Inc. SAP is the first company to make it feasible for large enterprises, she said.
That changed thanks to two factors, according to Sallam. The cost of RAM fell significantly, making in-memory cost effective. Hardware vendors have also moved to a 64-bit architecture.
“The 32-bit architectures didn’t really support broad of use of in-memory for enterprise use cases,” she said.
Both factors led to smaller vendors introducing in-memory applications that were suitable for small and midsize businesses or workgroups, Sallam said. For example, one of the applications combined in-memory capabilities for faster performance, along with easy-to-use BI authoring tools and interactive analysis, she said. Users were able to associate different dimensions and models without having to build aggregates like with traditional online analytical processing models, one of the key advantages of in-memory computing.
Columns and rows
Within HANA, data is organized in both columns and rows in order for the data to be retrieved as efficiently as possible, according to SAP.
While row storage provides advantages in metadata storage and logging used in transactional type workload, according to the company, columnar data storage is considered “key” to efficient in-memory computing in that it has the ability to significantly compress data and that with columnar storage, only data that is directly relevant to execution is accessed.
SAP BusinessObjects and in-memory applications
Developed as a replacement for disk-based relational database management systems, SAP has said that it considers HANA and in-memory technology the centerpiece of its IT strategy going forward, and that it will eventually power all SAP applications.
While SAP equates HANA with being the “engine” that churns and processes the data, business intelligence (BI) software has to sit on top of HANA to interpret the data. Although other BI tools will work, SAP says HANA was built to work and integrate closely with SAP BusinessObjects.
“That is probably the most common use case in terms of having some kind of analytical tool sit on top of HANA. That’s because everything is in the family, we’ve optimized the SAP business intelligence suite, to take advantage of and gain access to HANA,” according to Dan Kearnan, director of SAP business intelligence.
“So, if you want to design a report, get some analysis, build a dashboard, and you want to make sure that you’re getting the data you need to from HANA, you can do that through the suite of SAP BusinessObjects business intelligence.”
SAP is also in the process of rolling out a line of prebuilt in-memory applications for HANA, some of which are completely new and some of which are in-memory versions of existing applications. Those applications are designed for a specific need in mind.
“It can be for an industry, for a gap, retail or CPG [consumer packaged goods] sector. It can be for a line of business, for an HR manager. It’s whatever are the top KPIs [key performance indicators] that they need to measure and understand on a day-to-day basis,” Kearnan said.
So far, SAP has generally released two in-memory applications, including SAP BusinessObjects Strategic Workforce Planning, which lets companies simulate organizational changes in real time and then see how the changes will affect business. With the application, personnel executives see how their companies’ acquisitions and entrances into new markets will affect organizational structure, according to SAP.
SAP has also released two more in-memory applications, including Smart Meter Analytics for utility companies and a profitability analysis accelerator, or CO-PA, for companies processing inordinate amounts of financial data. SAP is in the process of introducing in-memory applications for cash and liquidity management, trade promotion management and sales and operations planning, among others.