HPE Storage Solutions for SAP HANA Data Tiering

[StorageExperts]

Enterprises face a major challenge today in terms of coping with explosive data growth being generated by their business applications. With SAP HANA, where data resides in memory, this presents an even greater challenge. Servers can only scale-up to a limited amount of RAM to accommodate data growth, making the cost of memory prohibitive. To keep all of the data in the RAM, IT manager find themselves faced with the only option of scale-out and adding more servers, which is cost-intensive and complex.

SAP tries to solve this problem by classifying its data into hot, warm and cold temperature ranges. Hot data is that which is frequently accessed by mission-critical applications; it needs to be highly performant and always available. Warm data is that which is less frequently accessed, and doesn’t have the same high availability and performance requirements as hot data. Lastly, cold data is that which is not accessed much and need not be part of the same HANA database. With this approach, data can be made to reside on a hardware tier corresponding to its access and performance requirements. This is the concept behind data tiering in SAP HANA.

This article is the first in a series that will take you through various SAP technologies for tiering data. We’ll also explore how HPE Storage enables efficient and easy implementation of these technologies.

Figure 1: Data growth in SAP HANA environment

Figure 1 shows an example of a data growth scenario for an SAP HANA database. Starting in 2020, the database consists of 2 TB. If no data tiering is planned and implemented, the database has the potential to grow up to 8 TB by 2026, and all of this data would need to be provisioned in memory. That means sufficient hardware would need to be provisioned upfront, or implemented in phases.

This scenario still means that you will face a large investment for expensive in-memory hardware infrastructure. If data tiering is implemented, however, this 8 TB of data can be divided into hot, warm and cold tiers, based on the business value and usage. If the amount of business-critical hot data roughly remains the same through the years, the compute resources for that data can remain consistent as well. Warm and cold data then can be offloaded to appropriate tiers. This approach can provide significant cost savings at the infrastructure layer, along with ease of management and predictability in performance.

 Which warm data tiering mechanism to use when?

Taking a look at the broader SAP applications portfolio, Figure 2 shows the relevant tiering technologies that are available, and the appropriate HPE Storage solutions that can be used to implement them.

Figure 2: HPE Storage solutions for various SAP data tiering mechanisms

SAP has three data tiering mechanisms. These are:

• Native Storage Extension. This essentially involves extending the native storage, i.e. the data volume on a HANA system, and offloading the warm data to it. This is applicable to native HANA scale-up deployments only as of HANA 2 SPS 4. HPE Primera, 3PAR and Nimble SSDs support this technology.
• Dynamic Tiering. This involves creating a separate commodity hardware tier and moving warm data from the HANA memory to this tier. Warm data resides on the disk in this tier instead of in memory. This method is also applicable only for native HANA scale-up deployments. HPE Primera, 3PAR and Nimble with SSDs and HPE 3PAR and Nimble with HDDs can be used to implement dynamic tiering.
• Extension Nodes. These are meant for scale-out native HANA as well as BW HANA systems. Here one node in the scale-out cluster is designated as an extension node, holding only warm data. HPE Primera, Nimble and 3PAR with SSDs can be used to implement extension nodes.

Future blogs in this series will talk about each of these technologies in detail, so stay tuned. Do drop in a comment on how these solutions fit you or your customer’s SAP environment and let us know if you have any questions.

Until next time - Happy tiering!

Regards,
Anshul

---

Meet HPE Blogger Anshul Nagori, Worldwide SAP Technical Marketing Engineer. Anshul works for the worldwide storage solutions team at HPE, and has more than 10 years of IT experience. He is a regular speaker at SAP and HPE events. His areas of focus include SAP HANA storage, data management, tiering, virtualization and data protection solutions.

Connect with Anshul on LinkedIn!

Storage Experts
Hewlett Packard Enterprise

twitter.com/HPE_Storage
linkedin.com/showcase/hpestorage/
hpe.com/storage