With last week’s conclusion of the World Series, it seems fitting to invoke legendary Yankees catcher Yogi Berra’s famous observation: “It’s like déjà vu all over again.” This sentiment perfectly captures the recent pattern of cloud outages, with multiple major disruptions occurring within a short timeframe.
Alan Gin, CEO and Co-Founder of ZeroDown Software, recently published a timely & insightful article titled “When the Cloud Fails: What the Amazon Outage Teaches Us About Real Resilience.” In it, he emphasizes a critical reality: cloud platforms, like any IT infrastructure, are not immune to downtime. As Alan notes, “The cloud does not equal continuity! The Internet’s biggest providers can (and will) fail.” His warning proved prescient when another major outage occurred just days later. According to an article from Yahoo it turns out that the latest outage originated from an “inadvertent configuration change” which caused Microsoft’s network and routing service to go down. What is interesting is that is that the outage is described as coming down to “interdependence and the widespread impact of the (Microsoft) outage.”
According to Insurance Times, the AWS outage alone resulted in estimated losses between $38 million and $581 million. The report states that “In terms of industry effects, we see those that depend on high availability being the most affected, including technology and financial services firms.” The financial impact of this week’s outage has yet to be quantified.
The Question of Accountability
Alan poses a crucial question in his article: “When a hyperscaler outage knocks out hundreds of businesses at once, who should be accountable—the cloud provider, the enterprise, or the insurer?”
Ultimately, it is the end-user enterprise that bears the greatest impact. Whether insured for business interruption or not, organizations suffer damage to their brand and reputation when critical systems fail. LCM Solutions Inc. believes the solution lies in a collaborative, proactive approach involving all stakeholders.
A Partnership for Resilience
ZeroDown® Software is a leading provider of Ultra High Availability™ solutions for hybrid and multi-cloud environments, and it is presently working with leading organizations across the industry to ensure uninterrupted operations.
With its deep roots within the global risk and insurance enterprise, LCM is proud to be partnering with ZeroDown in promoting its revolutionary Always Available™ architecture. ZeroDown’s technology perfectly aligns with LCM’s core mission to Predict, Prevent, and Protect when it comes to business risk.
How ZeroDown Works: A Technical Overview
Regardless of where applications and data are hosted, organizations remain ultimately responsible for the safety and integrity of their information. Production data must be protected for short-term recovery (addressing accidental deletion or corruption), long-term disaster situations (handling site-level failures), and archival purposes (meeting regulatory compliance requirements).
While infrastructure backup solutions can protect configuration data, mission-critical business applications require an additional layer of protection. The ZeroDown solution offers Ultra High Availability (HA) for specific workloads by matching transactions across on-premises instances and any number of parallel systems. For example, ZeroDown can mirror applications, data, and transactions from one instance to another—whether on-premises, in the cloud, or across multiple cloud providers.
Invisible Failover
When an outage occurs, it becomes invisible to the end user. The user’s session simply moves directly to the recovery instance with full transaction integrity maintained throughout the transition.
The Technology Behind Always Available™
The ZeroDown solution utilizes a transaction journal synchronizer and software switch to ensure instant, transaction-accurate recovery during any outage. Architecturally, the solution is based on an active-active model where both primary and secondary instances are actively engaged in mirroring. This contrasts sharply with the common master-slave architecture used by most failover solutions.
All servers are “hot” and “active” with no server hierarchy and, consequently, no single point of failure. If one instance fails, the others continue to function with no visible customer impact. This approach goes beyond standard disaster recovery by keeping applications available and operating at full efficiency, regardless of unplanned outages arising from hardware failure, human error, or natural disaster.
Transaction-Based vs. Block Replication
Unlike traditional disaster recovery (DR) solutions, ZeroDown® Software focuses on application transactions rather than block replication. Instead of moving large blocks of data across instances, the solution moves only transactions, which are significantly smaller in terms of data volume. The software installs as a Virtual Machine image that replicates the simultaneous processing of all transactions between application deployments. Transactions are independently processed at all sites, providing true active-active infrastructure.
The Path Forward
“For years, building applications that could seamlessly failover between AWS, Azure, and Google Cloud required enormous investment in parallel infrastructure. That’s no longer the case with ZeroDown SafeHouse™. Our Always Available™ architecture eliminates that complexity—keeping applications live, synchronized, and transaction-accurate across clouds, without the cost or risk of traditional redundancy.
LCM’s Predict, Prevent, Protect model aligns perfectly with this mission. We predict where vulnerabilities exist, prevent downtime through intelligent design, and protect operations by distributing workloads across hybrid and multi-cloud environments. Clouds will go down—your business operations shouldn’t. With ZeroDown Software, making modern applications resilient is no longer an impossible task—it’s simply how resilience should be built.”
— Jeff Edwards, Chief Evangelist & EVP, ZeroDown Software
