A regional cloud operator running exclusively in Southeast Asia and Australia, with datacentres in Singapore, Jakarta, Sydney, and Ho Chi Minh City. Meridian Grid emphasises in-region data residency, local-currency billing, and low latency for APAC end users, offering standard compute, object storage, and a managed Kubernetes service.
Share your experience to help others choose the right provider.
We serve mostly Indonesian users and moving from a Singapore-only setup to their Jakarta zone cut our p95 by around 40ms. That's not nothing for a mobile app. Data residency paperwork was also straightforward, they've clearly been through it before with regulators.
Being invoiced in SGD with local tax handled properly saves our finance team a genuine amount of hassle, and we're not eating FX spread every month. Small thing that nobody markets well. Platform is solid, k8s is a bit behind upstream but stable.
The docs say S3-compatible. In practice multipart uploads over a certain size behave differently and a couple of standard libraries choke on their ETag format. We spent the better part of a week working around it. Everything else has been good but I'd urge anyone to test their exact tooling before committing.
Half the instance families available in Singapore aren't in Sydney, and the GPU options are nonexistent there. We ended up splitting workloads across two regions which complicated our networking more than I wanted. Support was honest that Sydney is smaller and gave a rough roadmap, which I appreciated even if the answer wasn't what I wanted.
Singapore zone had a power event last year, about 90 minutes for us. They communicated well throughout, which is more than I can say for previous providers. Autoscaling works but is slower to react than I'd like — takes about 4 minutes to bring new nodes into the pool.
We serve mostly Indonesian users and moving from a Singapore-only setup to their Jakarta zone cut our p95 by around 40ms. That's not nothing for a mobile app. Data residency paperwork was also straightforward, they've clearly been through it before with regulators.
Being invoiced in SGD with local tax handled properly saves our finance team a genuine amount of hassle, and we're not eating FX spread every month. Small thing that nobody markets well. Platform is solid, k8s is a bit behind upstream but stable.
The docs say S3-compatible. In practice multipart uploads over a certain size behave differently and a couple of standard libraries choke on their ETag format. We spent the better part of a week working around it. Everything else has been good but I'd urge anyone to test their exact tooling before committing.
Half the instance families available in Singapore aren't in Sydney, and the GPU options are nonexistent there. We ended up splitting workloads across two regions which complicated our networking more than I wanted. Support was honest that Sydney is smaller and gave a rough roadmap, which I appreciated even if the answer wasn't what I wanted.
Singapore zone had a power event last year, about 90 minutes for us. They communicated well throughout, which is more than I can say for previous providers. Autoscaling works but is slower to react than I'd like — takes about 4 minutes to bring new nodes into the pool.
Signed up from Europe because the pricing looked good on paper. Obviously the latency is awful from here, entirely my own fault, but also support hours meant a 12 hour turnaround on everything. Cancelled after a month, refund was processed without argument. Rating reflects my experience, not the product's quality for its actual market.