Egress Charges and the Public Cloud: It’s Complicated!

In our technology consulting practice, we encounter customers who’ve had painful experiences with unanticipated costs in the public cloud.

Many of the same customers are being approached by MSPs who want them to keep hardware and computing workloads in their colocation facilities. MSPs generate custom cost-calculations that meet or beat equivalent services in hyperscale vendors such as AWS, Azure and GCP. They hope to convince companies to take advantage of metered services without changing their hardware paradigm and shifting everything to the public cloud.

These calculations and comparisons may seem straightforward but can result in even more hidden costs. Most organisations do not have a deep understanding of the many financial implications a complex digital transformation process can have on company finances.

Take the example of calculating the cost of Egress charges for an organisation that decides to implement Backup as a Service (BaaS) or Disaster Recovery as a Service (DRaaS) of their colocation infrastructure to a public cloud provider such as AWS or GCP.

Most cost modelling engines provided to customers by public cloud vendors and MSPs hosting their infrastructure will ask a series of questions to determine: 1) the storage tier required, and 2) whether the stored data will leave the public cloud, triggering a one-time Egress charge.

IT operations teams are generally familiar with the SLAs and their requirements for RTO/RPO and regulatory and compliance requirements which dictate the frequency of restores back to on-premises hardware. But the same IT ops teams are unlikely to be involved in discussions about Egress costs.

Specifics of the BaaS application can add further complexity. Increasingly popular B2OS (Backup to Object Storage) solutions often split data objects into multi-part objects that are written to the public cloud storage location. This data splitting can result in a massive increase to Egress fees based on the complex metering structure in Public Cloud. Even more confusing, hyperscale environments often make metering cost-favourable in one component of the data path while making it more expensive in another.

PTS understand all these complexities so you don’t have to. We’re here to ensure your applications, use-cases and architecture are optimised, now and in the future. We have the expertise to perform the deep-level analysis your digital estate deserves, and to help you make the best decisions going forward.


Our team is raring to go! Let us know a bit about yourself and your requirements and we’ll get you booked in for your free workshop.

Free Workshop