When your dev group tells you they’re planning to maneuver one in every of your key purposes to the cloud to turn into a SaaS service, often one in every of two issues will occur:
- First, they may try and recreate the precise performance with the identical structure working on rented computer systems within the cloud. Advice: Run away, the challenge is doomed.
- Second, they may reimagine and re-engineer that software to benefit from a few of the elementary advantages of cloud computing; in different phrases, the cloud-native strategy.
While it’s attainable to port conventional purposes and knowledge operations on to the cloud, many SaaS distributors have gained first-mover advantages from embracing a cloud-first, cloud-native structure. In this version of eWEEK Data Points, George Demarest, Senior Director of Marketing at Kyligence, explains why he believes the cloud-native strategy will win out over a direct software program port to the cloud.
Data Point No. 1: Elasticity gives a price benefit
It is sort of by no means a good suggestion to aim to match your bodily infrastructure specs to cloud digital infrastructure. Even although Amazon, Azure, and Google clouds present digital machines of all sizes that kind of match the specs of their bodily counterparts, you’ll nearly by no means be getting the very best deal to your cloud spend. These platforms present versatile pricing choices like “reserved instances,” enterprise agreements, and financial savings plans. Your cloud admin in IT can information you.
Data Point No. 2: Separation of computing and storage enhances each
In your knowledge middle, the servers you buy usually have some direct-attached storage (DAS) that you should use to retailer short-term information, photos, paperwork or no matter. But, whenever you enter the SaaS area within the cloud, it’s harmful to depend on this mannequin as a result of your laptop/CPU wants could rise and fall very in another way than your knowledge storage wants. The cloud lets you use object storage providers corresponding to AWS S3 or ADLS, which might be bought, optimized and managed individually out of your computing necessities. This separation of computing and storage will assist you in avoiding a “success crisis”–like including 10,000 new customers.
Data Point No. 3: Scaling reads and writes individually scales each
Similarly, when you find yourself deploying data-rich SaaS providers to a doubtlessly large concurrent person base, it’s possible you’ll need to select the very best knowledge discovery, knowledge manipulation and knowledge retrieval applied sciences. In the previous, relational databases might need been the logical selection for these capabilities, however at cloud-scale knowledge volumes and customers, it could make sense to choose extra specialised cloud providers, like columnar storage, in-memory databases, or knowledge streaming. This means, if a majority of your workloads are read-intensive and your database writes are bursty or intermittent, then your regular SaaS operations proceed, even when writes could spike (say, on the finish of 1 / 4 or yr). This can present a greater person expertise and a extra resilient working mannequin.
Data Point No. 4: Design for cloud object storage
Going additional with the problem of storage, a cloud-first design determination is to deal with the benefits supplied with cloud storage providers, corresponding to S3 or ADLS. Cloud suppliers will likely be beneath aggressive stress to enhance and innovate inside their storage providers. Application architects who carefully observe and shortly adapt to those improvements could have each benefit over opponents who’re extra circumspect. Take, for instance, Amazon’s current addition of read-after-write consistency. Having this characteristic constructed into the storage could imply that paying for some kind of SQL question engine will not be essential for some use circumstances. Other areas that…
