by Paul Speciale, chief marketing officer, Scality
Big data presents amazing opportunities for businesses to get valuable insights – but only if you can keep up with explosive data growth and demands for faster data access. Therein lies the challenge – but it’s not insurmountable. The key is to collect valuable benefits from a broad range of applications and data formats, including the latest AI/ML advancements.
The fastest-growing data type, unstructured data, may be optimized by IT professionals using Scality in a secure, on-premises environment that offers a cloud-like experience The insights gained from massive amounts of data managed with Scality help with making business decisions.
The challenge: Bringing together Big Data
Data analytics tools can provide your company with the knowledge needed to make strategic decisions about new products or services, business expansion opportunities, improved security and more. But how can the cost and accessibility of managing huge databases be improved? After all, the public cloud may not be as budget-friendly as we were led to believe.
The advantage: object storage
Scality’s object storage offers unlimited scale for data lakes and analytics applications–in multiple dimensions – including any number of files or data objects, from any number of sources, to feed many concurrent workloads. Its unbreakable cloud storage for data centers protects data and preserves availability so organizations can conduct data analysis at whatever scale necessary. This provides the advantages of public cloud but with the control, performance and security of on-premises private cloud infrastructure.
A use case to learn from
Let’s look at an example of how Scality RING, working with Splunk, manages Big Data analytics on a massive scale. A European travel service provider works with some of the biggest European airlines and serves a vast volume of travel searches each day. It targets an in-memory database for the required performance.
The logs of all these searches need to be captured and analyzed for patterns – where travelers go and when – to help determine pricing. The company needed a solution for the resulting 1PB of log data per day and one that could provide access to Splunk analytics.
Scality RING fits the bill, serving as the production storage for this data. The daily petabyte of data is stored as a 14-day rolling history, with the oldest (the 14th day) one PB pruned each day. RING has the ability to perform these actions while allowing 300 Splunk indexer nodes to read two weeks’ worth of data and perform peak throughputs to 60 GB/second.
This represents one of the largest Splunk deployments: 300 Splunk indexers on 36 HPE Apollo 4200 servers.
Tech partner ecosystems make it possible
Our extensive technology partner ecosystem (130+ ISVs and some of the largest infrastructure vendors) – help make such large-scale projects possible. That includes the likes of Vertica, Splunk, Elastic and Weka, to name just a few.
Scality has deployed hundreds of customer solutions, representing well over an exabyte of storage. For organizations that need an on-premises cloud storage solution for modern data-intensive applications, Scality provides object storage for data archives and tiers behind primary storage arrays.
RING offers on-premises storage scalability that is comparable to that of the cloud. RING can scale to host an infinite number of objects and thousands of server nodes spread across several sites.
Scaling on demand
As data continues to proliferate at lightning speed, organizations must have the ability to use it to its best advantage for the sake of the business. This requires analysis of massive quantities of data and the ability to scale data storage as needed. Working with our partners, Scality has created object storage solutions that help customers achieve these goals.
A different version of this blog was previously published with our partners at Bynet. To read that piece, click here: https://www.bynet.co.il/blog/big-data-analytics-challanges/