Intelligent Data Centres Issue 61 | Page 32

“ SECURITY OF
BOTH DATA AT REST AND DATA IN TRANSIT MUST BE CONSIDERED – DATA MOVING BETWEEN DATA CENTRES MUST NOT BE UNNECESSARILY EXPOSED TO THREAT VECTORS .
E D I T O R ' S Q U E S T I O N

STEVEN JACQUES , CONSULTING ENGINEER , JUNIPER NETWORKS

The key elements around regulatory and even legislative changes – as they relate to security and resilience – are the threat of greater punitive consequences and deeper requirements for threat incident reporting and response . There are multiple aspects to consider for compliance with these regulations .

Firstly , any security posture must mirror the corresponding resilience architecture . For example , if secondary or cloud-

“ SECURITY OF

BOTH DATA AT REST AND DATA IN TRANSIT MUST BE CONSIDERED – DATA MOVING BETWEEN DATA CENTRES MUST NOT BE UNNECESSARILY EXPOSED TO THREAT VECTORS .
This would typically mean common central orchestration and management systems , allowing security postures at each site to be synchronised . Security of both data at rest and data in transit must be considered – data moving between data centres must not be unnecessarily exposed to threat vectors .
Secondly , every connection should be thought of as carrying a potential threat . This means connections from corporate sites into the data centre , from data centre to data centre , between rooms in a single site , or even between virtual functions on the same host , are all candidates for security policy control and inspection . This concept underpins the Zero Trust model – security at the edge of the data centre alone is no longer sufficient .
Thirdly , data centre security designs must embrace innovation . Attackers already leverage cutting-edge technology such as Quantum Computing and AI – defenders should do likewise . AI will be increasingly central to both threat mitigation and response – consider AI models able to quickly detect zero-day malware , or the use of AI operations ( AIOps ) for threat forensics , which can already do tasks in seconds that a human operator may manage in hours ( and may miss something important ).
Finally , there should be increased emphasis on forensics and security operations . This is an important aspect of new regulations and while many data centres often have strong security mitigations in place , capabilities to analyse , respond and report on detected threats are often overlooked . AIOps can certainly help here , but this is perhaps the single area where automation is most important – operators may have vast quantities of network and threat data to examine , so automating associated processes can prevent exceeding regulatory timeframes for resolution and reporting . based backup data centres are used , access policies and threat mitigation systems must be identical at each site .
32 www . intelligentdatacentres . com