GlassFish Cluster Hosting on MassiveGRID PaaS
The Auto-Scalable GlassFish Cluster deploys a ready-to-work Eclipse GlassFish cluster on MassiveGRID PaaS: one Domain Administration Server (DAS), two worker instances and an NGINX load balancer, with auto-scaling triggers that add workers under load. Deploy a Jakarta EE application once and it runs on every instance.
GlassFish Cluster on MassiveGRID PaaS installs in one click in New York, London, Frankfurt or Singapore, costs about $63.96 a month for the Starter configuration, a Domain Administration Server (1 GiB), 2 × Worker instance (1 GiB each) and a NGINX load balancer (256 MiB), and starts with a 14-day free trial. See the sizing table.
Both buttons open the MassiveGRID PaaS dashboard with GlassFish Cluster pre-selected. Sign up without a card, name the environment, pick a region, and the package installs in a few minutes.
GlassFish Cluster Package Facts
| Attribute | Value |
|---|---|
| Package | Auto-Scalable GlassFish Cluster |
| Type | Application (installs a complete environment) |
| Category | Clusters & Containers |
| Software vendor / project | Eclipse Foundation (Eclipse GlassFish) |
| Licence | EPL 2.0 / GPLv2 with Classpath Exception (open source) |
| Runtime / stack | GlassFish DAS node, worker instances, NGINX load balancer |
| Regions | New York, London, Frankfurt, Singapore |
| Billing | Pay-per-use, about $63.96/month for the Starter configuration (26 cloudlets at $0.003372 per cloudlet-hour), no licence fee |
| Free trial | 14 days, no credit card |
| Uptime SLA | 100% |
What is GlassFish Cluster?
GlassFish is the reference implementation of Jakarta EE (formerly Java EE) and a full application server with EJB, JPA, JMS, JAX-RS and web profile support. Its clustering model uses a Domain Administration Server to manage a group of instances that share deployed applications and configuration, with in-memory session replication so users are not logged out if an instance fails.
Configuring a GlassFish cluster manually involves the DAS, node agents or SSH nodes, instance creation, load-balancer configuration and health checks. The package does all of that and adds platform auto-scaling: new worker instances are created, joined to the cluster and added to NGINX automatically when load rises.
What the GlassFish Cluster Package Deploys
Marketplace packages are not bare container images. One click provisions a complete, production-shaped environment on MassiveGRID infrastructure. How these nodes are sized and what they cost.
- Domain Administration Server: One DAS node hosting the admin console and cluster configuration, where you deploy applications to the whole cluster.
- Worker instances: Two GlassFish instances by default running your applications with session replication, scaled horizontally on load.
- NGINX load balancer: A balancer distributing requests across instances with sticky sessions and health checks.
- Auto-scaling triggers: Rules that add and remove worker instances and keep the DAS and balancer in sync.
What GlassFish Cluster Costs to Run
Three starting configurations for the GlassFish Cluster package, sized in cloudlets (1 cloudlet = 128 MiB RAM + 400 MHz CPU) and priced at the published rate of $0.003372 per cloudlet-hour, or $2.46 per cloudlet-month.
| Configuration | Topology | Cloudlets | Est. per month |
|---|---|---|---|
| StarterDevelopment | Domain Administration Server (8 cloudlets, 1 GiB), 2 × Worker instance (8 cloudlets, 1 GiB each), NGINX load balancer (2 cloudlets, 256 MiB) | 26 | $63.96 |
| StandardProduction | Domain Administration Server (8 cloudlets, 1 GiB), 2 × Worker instance (16 cloudlets, 2 GiB each), NGINX load balancer (4 cloudlets, 512 MiB) | 44 | $108.24 |
| ProductionHigh traffic | Domain Administration Server (8 cloudlets, 1 GiB), 4 × Worker instance (16 cloudlets, 2 GiB each), NGINX load balancer (4 cloudlets, 512 MiB) | 76 | $186.96 |
Estimates assume the environment averages the listed cloudlets around the clock; billing is hourly, vertical auto-scaling lowers the figure when the application is idle, and volume discounts apply as usage grows. Disk and external traffic are billed per gigabyte on top. No licence fee applies to the package. Pick a row in the installer by setting the cloudlet limits per node, or start small and let auto-scaling grow it. Price a different topology or a working-hours schedule in the PaaS pricing calculator; the PaaS page describes the billing model.
Why Teams Choose GlassFish Cluster
Full Jakarta EE
EJB, JPA, JMS, CDI, JAX-RS and JSF with the reference implementation.
Deploy once, run everywhere
Applications deployed via the DAS are pushed to every instance, current and future.
Session failover
In-memory session replication keeps users logged in when an instance is lost or scaled away.
Elastic capacity
Worker count follows load, so a quiet application runs two instances and a busy one runs more.
How to Install GlassFish Cluster on MassiveGRID PaaS
Installing the package takes three steps and a few minutes of waiting. No servers to provision, no configuration files to edit.
Open the installer
Click Install with free trial or Install to my account. The MassiveGRID PaaS dashboard opens with the GlassFish Cluster package pre-selected, so nothing has to be copied or pasted.
Sign up or sign in
New to the platform? Sign up for the free 14-day trial, which needs no credit card. Existing customers sign in with their usual dashboard credentials.
Choose region and deploy
Name the environment, pick New York, London, Frankfurt or Singapore, and confirm. The platform builds every GlassFish Cluster node, wires them together and hands you the live URL and credentials.
Who Runs GlassFish Cluster on PaaS
Enterprise Java applications
Line-of-business systems built on Jakarta EE that need horizontal scaling and failover.
Modernising legacy Java EE
Moving applications from ageing on-premise GlassFish or Oracle servers to elastic infrastructure.
Education and evaluation
Teams learning Jakarta EE clustering on a real topology without hardware.
Why Run GlassFish Cluster on MassiveGRID PaaS
Built for uptime. GlassFish Cluster runs on MassiveGRID's high-availability platform, where every environment is covered by a 100% uptime SLA. Nodes live on redundant hosts, storage is replicated, and failed containers are restarted or migrated automatically.
Pay for what you use. Billing is hourly and per cloudlet (128 MiB RAM plus 400 MHz CPU). Vertical auto-scaling adds and removes cloudlets as load changes, so a quiet GlassFish Cluster environment costs a few dollars a month and a busy one grows without a migration. Full details are on the PaaS page.
Your choice of jurisdiction. Deploy in New York, London, Frankfurt or Singapore and keep data inside the region your users or regulators require. MassiveGRID is ISO 27001 and ISO 9001 certified and GDPR compliant, and has run hosting infrastructure since 2003.
No lock-in. The package is a standard Jelastic Packaging Standard (JPS) manifest. You keep root-level access to the containers, can export the manifest, and can move the workload to any other Jelastic-compatible platform or to your own servers at any time.
Related on massivegrid.com: Platform as a Service.
GlassFish Cluster Questions
GlassFish Cluster Guides from the MassiveGRID Blog
Scenario-driven guides on running GlassFish Cluster on MassiveGRID PaaS, written for the developers, operators and buyers who deploy it. Browse the blog.
More from the Marketplace
Packages that are often deployed alongside GlassFish Cluster, or that solve the same problem a different way. Browse all 83 packages.
Deploy GlassFish Cluster Today
Start the free 14-day trial and have GlassFish Cluster running in minutes. No credit card required.