Carol Evenson en Cybersecurity, Cyber Security, information security 20/9/2016 · 2 min de lectura · +700

Protecting Big Data for Your Company

Protecting Big Data for Your Company

With the growing demand for big data comes the growing demand to secure it. Network solutions such as firewalls or employee security awareness are still valid, but traditional database solutions such as permission levels or encryption are much less valid when you're talking about enormous data sets serving a specialized purpose.

Growing Datasets

Securing your big data archives is more of a challenge than securing your traditional databases. For one thing, obviously, it's bigger, and with today's multiple channels of data capture both on and offline it's only going to get a lot bigger, much faster. Setting aside for a moment the performance issues resulting from security measures such as encryption, there's significant impact on disaster recovery when something goes wrong.

Disaster Recovery

Big data protection policies must take into account the size, cost, and efficiency of backup-recovery solutions, as well as the need for frequently loading masses of historical data into your business intelligence models. Full data backups require too much time, and disk volumes may not be adequate to storing petabytes of information, especially considering 48% of data is semi-structured or unstructured. That's a lot of hardware when you think about maintaining copies of that huge data archive.

Ineffective Software

Neither are software solutions effective on these huge data sets, as running scans of all that diverse information for injection attacks or malicious code is also time consuming when it's done on a regular basis. Replication and encryption of data also eats up already-strained resources and affects performance. Blue Coat System's Facebook page can suggest some excellent solutions, but most companies aren't even thinking security issues when they think big data.

Persistent Data

Video and audio files may lie forgotten for months or years before new issues bring a need for them. Photos and documents on discontinued products may suddenly become useful again if upper management decides the time is right for engineering an upgraded version. While aging data becomes less relevant to analytics, there may come a situation where it's needed. Archiving aged data to other data stores doesn't eliminate the hardware and performance issues, but, when that data is needed, compounds it.

Infrastructure

Many organizations recognize the value of big data but simply can't keep up with it, or the security measures it entails. By 2018, the US could face a shortage of up to 190,000 analytics experts as more companies of all sizes seek to leverage the data they're amassing. The need for information security professionals grows along with it; according to the Bureau of Labor Statistics, at a rate of 18% per year. The challenges of storing, accessing, and protecting a growing mountain of information only becomes more difficult. Cloud storage from a growing number of vendors help solve some scaling issues, but both cloud vendors and in-house solutions are faced with the possibility of needing near-infinite storage in the next ten years that data storage technologies are unlikely to keep pace with as more real-time data is being constantly collected and stored for use.

Instead, data protection must be embedded in the archive itself, as data is written. Security solutions that address updates as they happen rather than volumes is more appropriate. More intelligent solutions are needed such as the platform from Blue Coat that employs real-time indexing and content inspection of network data that can send alerts when anomalous packets are detected.

When you're dealing with massive petabytes of data, whether it's maintained on one huge machine or across a complex network file system, you're looking at a potentially massive drain on resources for any high-volume operations, including disaster recovery and anti-malware solutions. While sensitive information like account numbers or social security numbers might be removed, you're still looking at a growing footprint that could represent more exposure to the theft of intellectual property or denial of service network attacks. Scrubbed data isn't safe data. More advanced solutions are needed.