Carhartt shifts old data to the cloud with Komprise

US-based clothing manufacturer Carhartt has saved 75% on storage costs for every gigabyte of data it holds by being able to monitor and migrate files from its on-premise locations to Azure cloud storage using data management software from Komprise.

The company uses Komprise to identify data that is beyond much likelihood of frequent use and is able to migrate it from on-premise storage at $1 per gigabyte to Microsoft Azure Blob and Cold storage at around $0.25 per gigabyte.

Michigan-headquartered Carhartt makes clothing at nine manufacturing plants in the US and Mexico, with satellite locations in Europe and the Far East. It has two primary datacentres with 3,200 staff and more than 1,000 contractors who access its systems.

It is primarily a Microsoft user, with SAP as its enterprise resource planning (ERP) provider.

Key issues that faced the company were a great deal of “data sprawl”, said systems engineer, Earl Williams.

“The main reason we ended up using Komprise was that we needed to keep the system trim,” said Williams. “We had 22TB [terabytes] on the main file server, with everything from HR information to user file shares. There was no rhyme or reason; there was just data.”

The file server runs a Windows 2008 R2 operating system and had “become a big mess”, according to Williams.

“If servers went down we wouldn’t have access to data and it meant we ran risks in terms of security,” he said. “The plan to implement Komprise was so we could get a grip on the box because it was getting out of control.”

So-called Komprise “observers” were deployed, which run as a virtual appliance at the customer site to analyse data across on-premise storage, move the data and provide transparent access to data stored in the cloud.

According to Williams, in Carhartt’s case most data is migrated off according to its age. It goes to Azure Blob, where automated incremental growth takes care of increasing volumes of data, and Azure Cold storage.

The company now has around 16TB of data on-premise and the same amount in Azure, which is 75% less costly.

Besides storage cost, Williams cites the benefit of ease of access for users to the company’s file servers. “It doesn’t take the user five minutes to open a file, and remote regions can access content on the Azure cloud,” he said.

Komprise is a storage analytics tool that can index content and migrate it according to user policy. It is based on the open source Elasticsearch search engine.

Komprise’s Deep Analytics functionality allows the customer to query potentially petabytes of unstructured data and create a distinct virtual data set from disparate stored data.

Customers can set policies to categorise data by when it was last accessed, who accessed it, type of data, and so on. Komprise also migrates to other storage, and can do so between protocols and to the cloud, with a stub left behind for local access.