Close

See How Your Peers Are Moving Forward in the Cloud

New research from CDW can help you build on your success and take the next step.

Dec 04 2024
Data Center

How to Harden Public Sector Data Centers Against Climate Risks

State and local government agencies are working hard to achieve cyber resilience, especially in regions most susceptible to natural disasters.

Data center backup and recovery have been top of mind in recent years for state and local governments, and not just because of ransomware.

In April 2023, Fort Lauderdale City Hall’s basement flooded after 26 inches of rain fell in 24 hours, knocking the government’s headquarters out of commission and causing IT outages. The city subsequently made it a priority to move servers out of buildings susceptible to flooding and power outages and into the cloud.

2023 set a record for having the most billion-dollar climate disasters in recorded history. In 2024, Hurricane Helene and Hurricane Milton caused $53 billion and $34 billion in damage, respectively. As climate risks facing data centers multiply, state and local jurisdictions must be prepared with backup and recovery solutions, regardless of whether their facilities are on-premises, in the cloud or hybrid.

Click the banner below to explore data center optimization.

 

Make Immutable Backups a Top Priority

Every agency needs a data backup. It could live in an on-premises facility that the jurisdiction owns, a colocation facility or in the cloud, but it needs multiple immutable copies over time. An air-gapped backup that cannot be altered for a preset amount of time is a lifeline in the event of a ransomware attack.

Synthetic full backups in particular are faster and more affordable than traditional full backups. You don’t have to use a synthetic full backup, but they can reduce bandwidth consumption and can be restored more quickly than traditional backups.

Data is the lifeblood of most digital systems and, at the end of the day, it’s the one aspect of your IT environment that you cannot get back if you lose it. Before you begin thinking about disaster recovery, devise a backup strategy that works well for your purposes, especially if you store data on your own servers.

WATCH: Lodi, California, adopted a new backup system.

Evaluate Disaster Recovery Models and Hardware Upgrades

Disaster recovery is the process of accessing your backups and the applications and systems you need to maintain operations in the event of a disaster.

As a rule of thumb, any DR site should be a plane ride away. The backup facility must be far away enough from the main site that a regional weather event or outage won’t render your backup environment useless.

Depending on your IT setup, there are different approaches you can take.

Data Center TOC

 

DR for On-Premises Data Centers

Most jurisdictions employ a hybrid cloud model, which means they have some on-premises infrastructure. It rarely makes sense to have a second on-premises site as a locality’s DR site. There are exceptions: Some major cities with large departments can make equipment and data center space available for DR purposes. Or maybe a county will have a facility perched high up on a hill and safely removed from flood risk with equipment to spare. However, these situations are rare and tend to put the DR site too close to the actual data center.

But even agencies that are wary of public clouds have warmed up to Disaster Recovery as a Service. Amazon Web Services, Microsoft Azure and Zerto all offer ways to maintain your on-premises presence while replicating to the cloud and failing over in the event of a disaster. The DRaaS model is significantly cheaper than managing a separate DR environment or standing one up in a colocation facility.

Hardware is another consideration for on-premises data centers. For instance, it’s crucial to update your uninterruptible power supplies’ battery on a three-year cycle. Having energy-efficient rack cooling can help keep the lights on longer in the event that you fail over to a UPS. DRaaS should be reserved for worst-case scenarios, but for local power outages due to storms, you should be ready with updated redundant power and cooling for on-premises facilities.

LEARN MORE: Modular data centers provide options for state and local governments.

DR for Public Cloud Infrastructure

One of the well-documented upsides of a cloud production environment is that it’s automatically distributed across multiple data centers, and that's before you even set up a multiple-region DR strategy within the cloud.

Even if you were to just lift and shift, you're unlikely to have a natural disaster bad enough to take down a whole region. What’s more, many cloud vendors feature built-in backups. Plus, you have the option to distribute your DR backups over different regions for added peace of mind.

Plan for a Rainy Day

Small cities, towns and counties are starting to shift over to cloud infrastructures to achieve the redundancy and availability needed for data resilience and business continuity. This is especially true if their tech stacks are primarily intended to serve nonemergency government employees.

But the cloud doesn’t work for all jurisdictions, especially large cities and state agencies that store sensitive information about residents and that require control over their facilities. If your jurisdiction is coastal or is in a region regularly beset by storms — and even if it isn’t — backup and recovery must be a top priority.

Whether you back up on-premises facilities with DRaaS or fail over to a colocation facility on the other side of the county, you must plan to protect your data and minimize downtime when that rainy day comes.

This article is part of StateTech’s CITizen blog series.

CITizen_blog_cropped_0.jpg

Dragos Condrea / Getty Images