Protecting financial institutions from downtime and data loss

Banks and credit unions face four key threats to business continuity plans: cyberattacks, systems failures, natural disasters and cloud outages.

In today’s digital economy, a few minutes of downtime for critical applications and databases needed for online banking can be devastating – loss of customer satisfaction, negative press and social media, drained IT resources, reduced end user productivity, etc.

Be aware of four key threats to financial services organizations when evaluating business continuity plans: cyberattacks, systems failures, natural disasters and cloud outages. In the face of these threats, which applications and databases would incur the greatest cost to your organization were they to go offline?

Review your applications with other questions in mind: Would losing this system reduce employee productivity or disrupt operations? Would losing this system increase the workload of your IT team? Added work for your IT team could add to labor costs and costly delays to planned projects.

Sign up for the free BAI Banking Strategies newsletter and get industry insights delivered to your inbox.

Other questions can reveal costs that may be harder to quantify, but are impossible to ignore. What would losing a customer-facing application cost in terms of customer satisfaction and reputation? Negative publicity or social-media standing? If this application or database is locked by ransomware, what will that cost in terms of public confidence? Similarly, what if downtime draws regulatory scrutiny?

Having used these questions to identify your most critical applications, consider the main threats they face and how best to protect them.

Cyberattacks

Banks and credit unions may face the challenge of protecting vital applications and data without dedicated cybersecurity experts on staff. There are some important steps every organization can take to improve cyber security, regardless of size or IT resources.

The cost of ransomware and other cyber threats justifies the investment in an expert audit of regulated data and any means of accessing it – including firewall weaknesses, routers, network access points, and servers and recommended countermeasures specific to each weakness.

Document and communicate policies about the acceptable use of the company’s computer equipment and network, both in office and at home. Include clear restrictions for accessing and downloading sensitive data to local laptops and PCs, use of network access points, wireless security and best practices to avoid email-borne threats.

And apply software solutions for protection – this includes workstation/laptop antispam software, as well as automated security systems that hunt, detect and manage defenses against threats throughout the system.

Hardware system failures

Component failure within your IT infrastructure – servers, storage, network routers, etc. – is inevitable. To mitigate the cost of failure, answer these three questions:

  • How much data can you afford to lose? This is your recovery point objective (RPO).
  • How quickly you need to restore operations? This is your recovery time objective (RTO).
  • What level of application availability do you need? (Annual uptime percentage)

Your most critical applications – those that require an RPO of zero, an RTO of just 1-2 minutes, and true high availability (HA) of at least 99.99% annual application uptime – can be protected against hardware failure through failover clustering. For less critical applications and data, a simple backup or archiving plan may suffice.

Failover clustering provides redundancy for potential sources of system failure. Clustering software monitors application availability and if a threat is detected, this software moves the application operations to a standby server where operation continues with minimal downtime and near zero data loss.

Disaster recovery

Some applications may need protection from disasters that damage the local IT infrastructure. For applications needing HA, the primary and standby cluster nodes should be geographically separated, but connected by efficient replication that can synchronize storage between locations.

Cloud outages

Cloud infrastructure does not automatically provide application-level HA or disaster recovery protection. Cloud availability service-level agreements apply only to the hardware, which may not ensure that an application or database remains accessible.

Like any computing system, clouds are vulnerable to human error, disasters and other downtime threats. HA clustering for applications in the cloud should be capable of failing over across both cloud regions and availability zones. Traditional shared storage clustering in the cloud is costly and complex to configure, and is sometimes not available. Use block-level replication to ensure the synchronization of local storage among each cluster node. This enables a standby node to access an identical copy of the primary node storage and an RPO of zero.

By assessing the criticality of the applications, databases and systems required to operate efficiently and calculating the real cost of downtime for these systems, banks and credit unions can invest time and resources wisely to mitigate those threats cost efficiently.

Ian Allton is solutions architect at SIOS Technology Corp.