Long after the argument has been settled and the whole matter should have been put to bed, there is a lingering belief that SAP HANA is "too expensive." The argument goes like this: whatever its advantages as a data platform (which, of course, are many) SAP HANA is a luxury. It is good for a few edge use cases, but it is too niche for the enterprise world. It encourages businesses to "put everything into memory," which is extravagant and unnecessary.
For years, we have been deconstructing everything that is wrong with these misconceptions. No, you shouldn't run everything in-memory. No, SAP HANA is not a luxury. Yes, SAP HANA is the right platform for the enterprise and it is anything but niche. In recent years, the case against the "SAP HANA is too expensive" argument has grown stronger and stronger. And now, with the introduction of SAP HANA Cloud, it is overwhelming.
Apples to Apples
The principal way that naysayers have framed their "too expensive" argument over the years has been to make apples-to-oranges comparisons between SAP HANA and other platforms. Comparing an SAP HANA deployment to, say, an Oracle or (more erroneously) a Hadoop deployment is off-base in several important ways. It's easy to show that it costs more to deploy an entire enterprise dataset in-memory with SAP HANA than it does to deploy the same dataset on disk with a conventional database. And the difference looks even starker when the SAP HANA environment is compared to Hadoop. But these kinds of comparisons miss several critical points:
- SAP HANA is architecturally distinct from conventional databases in several important ways (and even more different from Hadoop.) Its data compression capability, coupled with its ability to reduce duplication of data throughout the enterprise, completely changes the economics of any comparison between platforms. SAP HANA helps to reduce the overall enterprise data footprint, rather than massively expanding it.
- It isn't just a question of how much it costs to store the data. It's a question of what you need to do with it. If you are running high-speed transactions and complex analytics that demand real-time results, Hadoop is not going to be able to deliver. An environment that "costs less" but that can't meet the fundamental needs of the business doesn't really represent any kind of savings. The conventional database might have a better chance of meeting those kinds of requirements, but to do so, it will almost certainly be required that the data be tiered between in-memory, high-speed disk access, and slower disk access.
- Through Dynamic Tiering, SAP HANA has been able to provide a tiered, multi-level, multi-temperature data storage environment for some time. The real comparison would be between such an environment supported by a conventional database and the same environment supported by SAP HANA. That comparison would have to take into consideration, as noted above, SAP HANA's data compression and other data-footprint-reducing features.
Removing these erroneous comparisons enables a valid analysis that takes into consideration SAP HANA's performance capabilities, its architectural advantages over other platforms, and its typical deployment (which includes a rational tiering of storage). In light of that kind of analysis, the "cost savings" of using a conventional database or other platform disappear.
Advantage: SAP HANA Cloud
And now, with a fully cloud-native version of SAP HANA available in SAP HANA Cloud, SAP has driven the final nail into the coffin of the "too expensive" argument. SAP HANA Cloud adds the advantages of cloud economics to the performance, scalability, and flexibility already inherent in SAP HANA. For starters, SAP HANA Cloud has brought the full multi-level, tiered storage architecture into the SAP HANA environment. Now users can manage the full tiered dataset from the cloud with SAP HANA Cloud. The top tier, for storing hot data is, of course SAP HANA itself — providing real-time, in-memory access to your most critical data.
The new Native Storage Extension provides the second tier, supporting warm data. It is a disk-store option for SAP HANA Cloud, providing an affordable alternative for near-real-time access to data that is important, but not as critical as the hot data. With the Native Storage Extension, users can store data on disk without having to integrate with a separate, external database. This provides seamless integration of hot in-memory data and your warm data on disk will be seamlessly integrated in a single SAP HANA environment.
Next comes the SAP HANA Cloud Data Lake, which supports cold data. It is a unique data lake environment, built on a database framework that can store large volumes of data for rapid access. In addition to cold data, the Data Lake can also serve as a repository for large volumes of incoming data such as sensor data from IoT (Internet of Things) systems. Because of the architecture, data stored in the Data Lake is ready to go for whatever analytical (or transactional) application needs to access it. Like the SAP HANA in-memory data platform, the data lake provides significant data compression and also helps to reduce unnecessary data duplication of the enterprise.
The complete integration between the tiers, from in-memory to data stored via the Native Storage Extension to data in the Data Lake, simplifies operations and reduces costs associated with integrating disparate systems. Reducing complexity and simplifying operations is one of the key advantages of SAP HANA Cloud over other cloud data management services. Amazon Webservices offers eight different databases and seven different storage systems; Google provides six different database offerings; Microsoft offers eleven different data management solutions. These choices put application developers or system architects in the same data management dilemma they faced in the last millennium: data resides in individual silos that are only capable of handling one specialized workload (say, transaction processing) and storage format. In order to perform additional processing tasks (analytical queries, for instance), data needs to be replicated and transformed into another system, bringing back the reporting delays and error-prone data extraction tasks modern database platforms were able to overcome. SAP HANA Cloud services largely eliminates these issues.
In addition to providing management of the full data environment, SAP HANA Cloud opens up possibilities that would never be achievable in an on-premises environment. In the cloud, increasing memory, storage, or computational capability does not require a permanent investment or commitment (as it would on-premises). A cloud environment can be scaled up temporarily to meet peak periods of demand, allowing the business to pay only for the additional capacity when needed. The cloud also opens up the possibility of scaling memory and CPU separately, enabling businesses to scale their environment with a precision never before experienced.
With the promise of these kinds of capabilities, SAP HANA Cloud presents a solution that is not only not "too expensive," but that can help businesses manage their assets more effectively both day to day and year over year. Far from being a luxury, SAP HANA Cloud can help ensure your business gets the highest possible return on investment from all its data assets.