The traditional ‘rear view mirror’ approach to analytics is no longer enough. Rather than drawing insights after the fact, analytics need to provide insights in real time to inform business decisions, or even predict future events before they occur. New technologies such as predicative modelling and AI are no longer adventurous options for exceptional businesses, they are crucial tools that need to be implemented across business verticals.
Luckily, a growing number of tools and hardware options for real-time insight are becoming available. In-memory analytics, which keeps data closer to the CPU, is one example. Some of its benefits include the ability to simplify infrastructure with more uptime and reduce latency for key applications. The upshot of these benefits is the delivery of faster, deeper insights at scale and in real time, which ultimately allows the organization to drive more revenue and enhance the customer experience.
One key to success in implementing in-memory analytics is strategically developing an IT infrastructure tailored to your business’ needs. Here are two examples of how different business considerations drive these infrastructure choices.
In-memory analytics can be used to power fraud detection and prevention in financial services. For example, when a transaction takes place, analytics can compare data about the transaction with historic transaction and customer data to determine how likely it is that the transaction is fraudulent. It is crucial that this comparison should happen almost instantaneously, so that the transaction can be either approved or declined.
There are several key stages to this process. First, live transaction data is received at the edge, where processing can begin. However, space constraints at the edge mean that the historic data may need to be stored in a central repository, where live data (once cleaned and filtered) is sent for processing. The data center requires a highly performant compute system, for example one delivered on Intel® Xeon® Scalable processors with Intel® Optane™ DC persistent memory - which, for example, delivers 2.4x better performance running SAP HANA* compared to a five-year-old system1 - to ingest all the available data and perform the comparison between historic customer data and the data from the new transaction. The data center then sends a signal back to the edge determining whether the transaction is approved or declined.
In the case of an organization diagnosing and treating cancer patients, there are additional important considerations, including protecting patients’ private information and complying with data regulations. With Intel Optane DC persistent memory, reduced restart time helps to deploy security updates more quickly to stay up to date.
Healthcare organizations create and process very large sets of data, such as magnetic resonance imaging (MRI) scans and other medical imagery, every day so an ability to scale quickly is also important. The IT infrastructure therefore needs to focus on meeting the requirements for memory and storage capabilities, as well as compute performance and network speed, which are essential to the processes of diagnostics and prognostics.
The system must also be able to process peaks when ingesting large volumes of new data. A CPU-based system with enhanced memory capabilities, such as the 2nd Generation of Intel Xeon Scalable processors run with Intel Optane DC persistent memory, can offer the capacity to scale up and run larger data sets when required.
By building their in-memory analytics capabilities on Intel® architecture, organizations can also benefit from the ability to move data through Intel® Ethernet products, store warm data with Intel® SSDs, and benefit from a hardware root of trust for more secure data handling. They can even integrate AI capabilities like machine learning and deep learning with software, toolkits and libraries that are optimized to run on Intel® technology.
To find out more about how to construct your infrastructure and run the best in-memory analytics for your organization, read the guide Build Your In-Memory Analytics Stack.
For more information, www.intel.com/yourdata
Follow @IntelBusiness on Twitter for more news.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors.
Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit www.intel.com/benchmarks.
Performance results are based on testing as of dates shown in configuration and may not reflect all publicly available security updates. See configuration disclosure for details. No product or component can be absolutely secure.
1 Testing by Intel IT March 12, 2019
Baseline: three-node (1-master + 2-slave) SAP HANA 2 scale-out configuration. Per Node: 4x Intel® Xeon® processor E7-8880 v3 (2.3 GHz, 150 W, 18 cores ), CPU sockets: 4; Microcode: 0x400001c, RAM capacity: 64 x 32GB DIMM, RAM model: DDR4 2133 Mbps; storage: GPFS, approximately 21.8TB of formatted local storage per node, SAN storage for backup space only; network: redundant 10GbE network for storage and access, redundant 10G network for node-to-node; OS: SUSE 12 SP2, SAP HANA: 2.00.035, GPFS: 22.214.171.124. Average time of 50 individual test queries executed 30-50 times each, for a total of approximately 25,000 steps: 2.81 seconds.
New configuration, one-node (1-master) SAP HANA 2 scale-up configuration: CPU: 4 x 2nd Generation Intel® Xeon® Platinum 8260 processor (2.2 GHz, 165 W, 24 cores), CPU sockets: 4; Microcode: 0x400001c, RAM capacity: 24 x 64GB DIMM, RAM model: DDR4 2133 Mbps; Intel Optane DC persistent memory: 24 x 126GB PMM; storage: XFS, 21TB; network: redundant 10GbE network; OS: SUSE 15, SAP HANA: 2.00.035, Intel BKC: WW06. Average time of 50 individual test queries executed 30-50 times each, for a total of approximately 25,000 steps: 1.13 seconds.
All information provided here is subject to change without notice. Contact your Intel representative to obtain the latest Intel product specifications and roadmaps.
Optimization Notice: Intel's compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.
Notice Revision #20110804
Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel.com
Intel, the Intel logo, Xeon, and Optane are trademarks of Intel Corporation or its subsidiaries in the U.S. and/or other countries.
*Other names and brands may be claimed as the property of others.
© Intel Corporation