Written by Mike Skrzypczak
Prior to the broad availability of cloud computing, organizations had little ability to scale their computing resources when necessary. For example, companies would need to buy computers, storage arrays, and other equipment, and find room in a datacenter for them. Such a commitment can be substantial, and months of planning would be required to budget for, acquire, and deploy resources.
Cloud computing, along with modern CloudOps practices, changed all that. Increasing capacity for certain kinds of applications like consumer-facing websites could now be done in minutes instead of months. As long as there were end-users using a web site, capacity could increase to meet that need, and then decrease with a drop in demand.
The number of users is elastic; they are the consumers that access the web site. The computing resources are elastic, so we should assume the output (for example, merchandise orders) of the ecosystem is elastic. Right? Simple?
Not always, especially when it comes to data analytics and reporting applications. Challenges exist with scaling database systems (that are designed for relatively steady transaction workloads) to accommodate workload spikes associated with analytics and reporting - limiting overall elasticity.
In the past few years, this problem has been addressed with horizontally scaling technologies designed for these kinds of workloads (for example, Apache Spark). However, scaling the compute-resources for analytics workloads is only part of the problem. The other is scaling labor required to create the reporting and analytics that ultimately extract value from an organizations data.
It’s often a backlog of data-related projects that lead organizations to Proxet for help, and we often recommend some sort of modern data platform. Deploying modern data platforms like Palantir Foundry, Databricks, Azure Fabric, and others — enhanced with industry standard data and interchange models like OSCRE — can go a long way to reduce the labor cost of data projects and make backlogs manageable.
At Proxet, we know the strategic value in modernizing data platforms is added capacity. A successful deployment gives time back to staff to self-service their own data projects, and gives the organization the elasticity to deal with the increasing demands for data-driven decision making.
Get in touch with us today to discover how we can help you enhance your organization's agility!
