The landscape of modern game development is evolving rapidly and becoming more complex than ever before. Studios are pushing creative and technical boundaries to deliver increasingly rich and immersive player experiences. However, this evolution brings significant infrastructure challenges. Managing sprawling codebases, massive libraries of high-fidelity assets, lengthy compile and build times, and coordinating distributed teams worldwide place enormous strain on the development pipeline.
Compounding these technical challenges are security concerns, especially when working with external partners, and the high financial costs of maintaining robust, high-performance infrastructure. Traditional on-premises environments struggle to keep pace. They lack the elasticity, automation, and scalability that today’s demanding workflows require. Moreover, shortages in IT resources, scarcity of skilled personnel, and global supply chain disruptions make on-premises infrastructure maintenance increasingly impractical.
This reality has driven a significant shift: game studios are moving to the cloud
Cloud platforms—particularly AWS—offer a vast array of innovation-enabling services, built-in resiliency, and the ability to dynamically scale compute and storage resources based on fluctuating developer and player demand. This shift unlocks opportunities to optimize game development end-to-end: improving performance, enabling seamless collaboration, and drastically reducing time to market.
But to truly harness the cloud’s potential, game studios need more than just scalable compute—they require a cloud-native file system that delivers enterprise-grade performance, security, and data management capabilities that they experienced with on-prem infrastructure.
Game Studios are Building with Amazon FSx for NetApp ONTAP
Amazon FSx for NetApp ONTAP is a fully managed AWS service that integrates AWS’s scalable infrastructure with NetApp’s industry-leading ONTAP file system delivered as a native AWS service just like AWS’s other native data storage services. It is purpose-built to empower developers to streamline and accelerate their entire development pipeline—from asset ingestion and code management to build automation and playtesting. Additionally, it can directly tackle core infrastructure challenges in game development, enabling studios to innovate faster, scale smarter, and reduce costs, which is a perfect fit for game development.
In this blog, we'll explore:
- The unique challenges of game development in the cloud.
- How Amazon FSx for NetApp ONTAP addresses these through real-world use cases:
- Game code repositories: How AAA studios use FSxN with Perforce Helix Core (P4).
- Game build pipeline hydration: How FSxN accelerates build hydration and reduces costs.
- Developer Virtual Desktop Infrastructure (VDI): How FSxN enhances workspace provisioning efficiency.
- How to start optimizing game development with FSx for ONTAP:
- Manual deployment via AWS Console
- Automation using the AWS Cloud Game Development Toolkit
- Advanced workload management with NetApp BlueXP Workload Factory
Bridging the DevOps Gap in Game Development with FSx for NetApp ONTAP
Traditional software development benefits from mature DevOps workflows supported by toolchains like Jenkins, GitHub Actions, Gradle, and Webpack. These tools thrive on standardized, repeatable cycles with large active communities.
Game development, however, operates on a different scale and complexity. Massive codebases and vast asset libraries are typically managed via Perforce P4, a version control system designed for scalability and performance. Once developers submit changes, continuous integration and delivery (CI/CD) systems trigger automated build and test processes that rely on ephemeral build agents. These agents spin up with the necessary tools, execute jobs, and terminate upon completion.
The build artifacts in game development—compiled code, 3D models, audio files, textures—are enormous and complex. Testing often extends beyond automation to include human playtesting, adding further resource demands. Unlike standard software, games target multiple platforms simultaneously—PC, consoles, mobile—requiring resilient, high-throughput infrastructure capable of handling massive workloads without bottlenecks or failure.
This complexity creates a DevOps gap—both on-prem and traditional cloud data infrastructure often cannot efficiently support the performance, cost, and flexibility demands of game development pipelines.
Amazon FSx for NetApp ONTAP is purpose-built to addresses this gap in the cloud for game development by providing:
- High throughput and low latency: Accelerate build pipelines by delivering shared storage that supports fast, concurrent access to instant thin clones of code depots across distributed build agents.
- Cost savings: The combination of advanced deduplication and thin cloning reduce data duplication and storage footprint for build pipeline hydration, yielding up to 70% savings on infrastructure spend.
- Data protection: Native snapshotting and cross-region replication improve backup and disaster recovery capabilities while greatly improving data resilience and availability for game data with native high-available architectures.
- Simplified management: Centralized control and consistency across applications and platforms reduce operational complexity.
- Global collaboration: FSx for ONTAP supports distributed teams by delivering high-performance, consistent, and centralized data access across multiple locations ensuring the data developers crave is where they need it when they need it.
- Scalability and automation: Flexible and elastic infrastructure that grows with your project demands, reducing operational overhead and can be automated via extensive and easily accessible APIs.
For game studios specifically, FSx for ONTAP becomes a strategic enabler—helping close the tooling gap while aligning infrastructure with the creative and technical realities of modern game production. To do this, FSx for ONTAP leverages:
- Unique Instantaneous Thin Cloning: Create writable, point-in-time copies of datasets instantly without full data duplication, ideal for test environments or rapid iteration. For game builds, this means creating near-instantaneous, space-efficient copies of large game builds and asset libraries, drastically reducing storage costs and eliminating unnecessary data duplication.
- Deduplication & Compression: Reduce redundant data storage and optimize disk usage.
- Enterprise Grade Data Protection
- Snapshots & Backup: Instant, space-efficient point-in-time copies support rapid recovery.
- Resiliency and security: Built-in snapshotting, replication, and cross-region capabilities protect your most valuable assets and minimize downtime.
- Efficient Cross-region Replication: Sync data between AWS regions or availability zones for disaster recovery while maintaining storage efficiency.
- Data Auto-tiering: Data is automatically moved between SSD and capacity tiers based on access patterns, balancing cost and performance.
Together, these capabilities significantly streamline game development pipelines by enabling faster iteration, lowering costs, and enhancing code protection. By reducing end-to-end build and test cycles from hours to minutes, teams can accelerate time-to-market and improve Mean Time To Resolution (MTTR), allowing for quicker, more frequent, and higher-quality releases. Advanced cloning and data efficiency technologies help cut storage costs by up to 70%, boosting return on investment (ROI) and lowering total cost of ownership (TCO). Meanwhile, robust security and resiliency features protect critical intellectual property, ensuring rapid recovery and minimal downtime in the event of a failure.
This can all be realized in three strategically targeted aspects of the complete game development pipeline:
- Game Code and Data Version Control Repositories
- Game Build Pipeline Hydration
- Game Developer VDI Workspace Provisioning
It's clear, game development is evolving - here's how some of the world's leading game studios are rethinking and optimizing core aspects of their game development pipelines with FSx for ONTAP at the center:
Use Case 1: Optimizing Game Code Repositories with FSx for NetApp ONTAP
When considering the game development process as a whole, creating and building a game starts with an idea made into code. Eventually, some of these games become core to critical success at their game studios necessitating the use of optimized solutions to manage the life of their award winning titles. Perforce P4 (formerly known as Helix Core) is the industry-standard version control system for game development, favored for its ability to handle massive critical codebases and complex branching workflows.
According to a 2023 TechValidate survey:
- 80% agree P4 is the most scalable, high-performance version control system
- 82% rate Helix Core’s scalability as "Best in Class" or "Significantly Better" than alternatives
- 100% reported ROI within 1-2 years of deployment
One of the major benefits of using Perforce P4 is its performance. P4 only preserves changes of updates in text files. For binary files, it compresses each version and saves them as chunks, known as versioned files. Even when a large development project increases the number of files or commits, the size of the repository (hxdepot) is not prone to explode. It also has a mechanism that works faster even if clients are physically away from the master server. Therefore, Perforce is used for version controlling not only text files, but also large binary files, such as images.
Modernizing P4 infrastructure by migrating these repositories and workspaces to the cloud is a natural step—but it brings challenges around cost, resiliency, and data flexibility due to the sheer size of game development data.
FSx for ONTAP extends the realized benefits of on-prem enterprise storage to Perforce P4 in AWS, enabling:
Flexibility, Scalability, and Elasticity as a well-architected AWS native service, FSx for ONTAP filesystems can be easily deployed in multiple flexible architectures for multiple P4 server types, scaled-out up to 72GBps of throughput, 2,400,000 IOPS, and ~1PB for a single namespace.
Furthermore, FSx for ONTAP can be easily and quickly adjusted to expand storage, throughput, and IOPS from within the AWS console, AWS and ONTAP CLI’s, various available APIs, or other available tools we’ll visit later in this article.
Up to 70% reduction in recurring storage costs via data deduplication, compression, compaction, and strategically leveraging FSx for ONTAP’s instant thin cloning in build processes. These efficiencies are maintained when leveraging FSx for ONTAP backups and replication for data protection. Data tiering can also be enabled for disaster recovery replicas that are infrequently used, even further reducing total data footprint.
Enhanced data protection and resiliency with improved RPO (Recovery Point Objective) and RTO (Recovery Time Objective) ensuring rapid recovery from failures or data corruption through:
- Flexible High Availability: Both Single-AZ and Multi-AZ deployments feature an HA pair of nodes that make a single filesystem. Single-AZ is a 2-node cluster in a single AZ and protects against a node failure in a single AZ. Multi-AZ is a 2-node cluster spread across 2 AZs and further protects from a single AZ failure.
- Advanced Snapshot Technology: Fast, space-efficient point-in-time backups which are also integrated into AWS and FSx Backups.
- Optimized Disaster Recovery: Cross-region replication (commonly known as SnapMirror on FSx for ONTAP), backups through integration with AWS and FSx Backup, and data tiering capability to balance cost and speed by tiering cold data to a cheaper and lower class storage medium.
Global availability and consistency supporting distributed teams with centralized control. Since FSx for ONTAP is a native AWS service, it is consistently available in multiple AWS regions capable of supporting multiple Perforce P4 server types and replication across all regions where FSx for ONTAP is available.
Critical high performance, low latency, and high throughput that game development hungers for. AAA studios have demonstrated significant performance gains by running Perforce on FSx for ONTAP in AWS by reducing latency and accelerating developer productivity while cutting infrastructure costs.
To achieve the highest performance possible, iSCSI (block storage protocol) on FSx for ONTAP for all P4 server volumes (/hxmetadata, /hxlogs, and /hxdepots) is considered the best practice. However, NFS can also be considered for the /hxdepots volume as a balanced performance vs. manageability option.
For a brief perspective on iSCSI vs. NFS performance differences, please consider the following analysis that leveraged Perforce’s general P4 multi-sync benchmark while testing multi-P4 client simultaneous multi-syncs:
For simultaneous multi-syncs up to 20 clients, the completion time was significantly better with iSCSI (block) storage. As we increased the number of clients to 150, the completion time was determined by network throughput, that is, the actual time it takes to pull the files into local client workspaces. In this case, iSCSI continued to outperform NFS with consistency.
However, regardless of which protocol is chosen for the /hxdepots volume, it is recommended to run multiple iSCSI or NFS nconnect sessions to achieve higher levels of throughput. This allows data to be transferred in parallel over multiple connections, thereby increasing the aggregate bandwidth and reducing the time it takes to complete I/O-intensive operations such as P4 syncs and submits. For example, the following configuration was used for the NFS testing shown above:
-o nfsvers=3,nconnect=8,local_lock=all
-n node.session.nr_sessions -v 8
Use Case 2: Accelerating Game Build Pipeline Hydration and Full Optimization
Hydrating build pipelines means populating the build environment with the latest game code and assets after changes are committed—critical for generating up-to-date builds for development, testing, and release.
Because game assets and build artifacts can reach terabytes in size, hydration can become a major bottleneck for game development studios. Copying large datasets consumes time and storage, slowing iteration and inflating costs.
FSx for ONTAP’s shared, high-throughput storage solves these problems by:
- Providing thin, instantaneous, and fully writable environments based on instant snapshots for rapid and cost effective provisioning.
- Creating new build environments in seconds with near-zero capacity overhead.
- Reducing build pipeline costs by up to 70% through thin cloning and data efficiency.
This results in:
- Faster build times and shorter MTTR.
- Enhanced collaboration with immediate access to the latest builds.
- Drastically lower infrastructure costs.
An award-winning AAA game development studio conducted its own analysis on cost savings between their current game build hydration configuration that leverages EBS vs. FSx for ONTAP leveraging their own game data for three distinct projects and achieved the following savings for Game Build Hydration with FSx for ONTAP:
While these cost savings are already drastic, FSx for ONTAP could be leveraged across the whole pipeline to reap further cost savings outside of just the build hydration portion of the pipeline as shown in the following sample architecture:
Here's how the sample architecture for optimizing the game build pipeline works:
- Build Trigger & Environment Provisioning
A build is triggered either manually via a Self-Service Portal or automatically via Perforce or code flag. Orchestration (e.g., Jenkins, TeamCity, Ansible, Terraform, Horde) provisions build nodes using a custom AMI (built with Packer) that includes all necessary tooling (e.g., .NET SDK, msbuild, compilers, gcc, etc.).
- "Syncer" Client Selection & Preparation
Orchestration monitors persistent lightweight instances called “Syncer” clients, which maintain replicated Perforce (P4) data from the Build-Edge. Each "Syncer" represents a unique workspace/changelist and mounts its primary volume from FSx for ONTAP, allowing persistent high-performance access and instant thin-clone capability. Once an available "Syncer" is found, it is synced to the latest requested changelist, enabling near-instant readiness due to the already-updated replica from Build-Edge.
- Volume Cloning for Build Execution
After the "Syncer" is updated, which only requires delta change updates, orchestration clones its volume using FSx for ONTAP thin clones, tied to the specific changelist. This clone is mounted to the already instantiated build node, providing access to the exact source code snapshot. Builds may read/write to this volume with minimal overhead due to snapshot-based delta management.
- Post-Build Output & Distribution
Build artifacts are written to S3 + CloudFront or back to FSx for ONTAP, depending on whether the use case involves developer collaboration (caching and testing), intermediate asset reuse, or external distribution to consumer platforms.
- Teardown & Reset for Next Build
Once the build is complete the builders are destroyed, their FSx for ONTAP clones are deleted, and the persistent "Syncer" clients remain, but are updated with the latest P4 data and reset for the next changelist request.
→ This cycle repeats efficiently for every new build request.
As shown, leveraging instant thin clones on FSx for ONTAP vs. other data services results in a drastic cost savings by reducing data requirements in the game build pipeline. Furthermore, because there is less data to copy, build hydration is significantly faster drastically reducing game builds.
Use Case 3: Optimizing Developer VDI Environments
Developers often require isolated Virtual Desktop Infrastructure (VDI) workspaces provisioned with the latest code and assets. Even further, these VDI workspaces need to travel with the developer, meaning the data needs to be globally mobile and consistent across regions. Traditional approaches use full snapshots, resulting in enormous data duplication and slow workspace creation.
Much like Build Hydration, because VDI orchestration would leverage similar code and instant thin cloning, FSx for ONTAP enables:
- Accelerated iteration: Workspace cloning from base volumes with FlexClone technology.
- Massive cost savings: Reduce copies from thousands of FULL snapshots to a handful of thin cloned volumes.
- Faster time to productivity: Reduces provisioning times from hours to minutes since FSx for ONTAPs cloning is a unique thin-provisioned cloning technology.
- Performance advantages: Provisioned file system-level performance outperforms individually provisioned native volumes.
A major AAA game studio analyzed that they would benefit from $4.4M of savings annually for their developer VDI farm and would expect to multiply these savings as their developer base grows by leveraging FSx for ONTAP cloning concepts. They reported:
- Drastic reduction in storage footprint.
- Near-instantaneous code updates via cloning and FlexCache, a native ONTAP feature for efficient caching at distance.
- Significant cost savings and operational efficiency.
How to Get Started
AWS Console Manual Deployment
Amazon FSx for NetApp ONTAP is a native AWS service, available directly through the AWS Management Console alongside other native storage options like S3 and EBS. This provides a simple way to deploy, configure, and monitor your FSx file systems.
AWS Cloud Game Development Toolkit (CGD Toolkit)
The CGD Toolkit is a modular collection of templates, automation scripts, and reference architectures designed to streamline game infrastructure deployments on AWS. The toolkit provides
- Guidance: Recommendations and best practices for common architectures and tooling, from Packer to Horde and even managing iSCSI on FSx for NetApp ONTAP.
- Assets: Individual playbooks, pipelines, and automation scripts for targeted tasks.
- Modules: Reusable Terraform configurations for deploying workloads.
- Samples: Complete reference architectures combining modules and AWS services.
For example, studios can utilize provided customizable Terraform and Ansible templates, purpose built with best practices, to deploy Perforce P4 on FSx for ONTAP using the CGD Toolkit automation, drastically simplifying deployment and setup of not only P4 and it's associated infrastructure, but also required FSx for ONTAP iSCSI storage as recommended previously in this blog.
Basically, turn this:
Into this:
module "unreal_horde" {
source = "modules/unreal/horde"
unreal_horde_service_subnets = aws_subnet.private_subnets[*].id
unreal_horde_external_alb_subnets = aws_subnet.public_subnets[*].id # External ALB used by developers
unreal_horde_internal_alb_subnets = aws_subnet.private_subnets[*].id # Internal ALB used by agents
vpc_id = aws_vpc.unreal_horde_vpc.id
certificate_arn = aws_acm_certificate.unreal_horde.arn
github_credentials_secret_arn = var.github_credentials_secret_arn
tags = local.tags
fully_qualified_domain_name = "https://horde.${var.root_domain_name}"
}resource "netapp-ontap_lun" "depots_volume_lun" {
count = var.storage_type == "FSxN" && var.protocol == "ISCSI" ? 1 : 0
cx_profile_name = "aws"
os_type = "linux"
size = var.depot_volume_size * 0.75 * 1073741824
svm_name = var.fsxn_svm_name
volume_name = aws_fsx_ontap_volume.depot[0].name
name = "/vol/${aws_fsx_ontap_volume.depot[0].name}/${aws_fsx_ontap_volume.depot[0].name}"
depends_on = [
aws_lambda_function.lambda_function,
aws_vpc_security_group_egress_rule.link_outbound_fsxn,
aws_vpc_security_group_ingress_rule.fsxn_inbound_link
]
}
resource "netapp-ontap_san_lun-map" "depots_lun_map" {
count = var.storage_type == "FSxN" && var.protocol == "ISCSI" ? 1 : 0
cx_profile_name = "aws"
svm = {
name = var.fsxn_svm_name
}
lun = {
name = netapp-ontap_lun.depots_volume_lun[count.index].name
}
igroup = {
name = netapp-ontap_san_igroup.perforce_igroup[count.index].name
}
depends_on = [
aws_lambda_function.lambda_function,
aws_vpc_security_group_egress_rule.link_outbound_fsxn,
aws_vpc_security_group_ingress_rule.fsxn_inbound_link
]
}**Disclaimer** The above snippets are only examples taken from: https://github.com/aws-games/cloud-game-development-toolkit/tree/main/modules/unreal/horde & https://github.com/aws-games/cloud-game-development-toolkit/tree/main/modules/perforce/
For more complete and comprehensive code with guidance, please visit the AWS CGD Toolkit here:
https://aws-games.github.io/cloud-game-development-toolkit/latest/
NetApp BlueXP Workload Factory
BlueXP Workload Factory automates workload lifecycle management on FSx for ONTAP—handling everything from initial design to ongoing operations with workload-centric, best-practice workflows. For Perforce workloads, you can create and mount clones of desired game versions for QA or release from directly within the Perforce P4 client itself. Code snippets are automatically generated of the automation used to create and mount the clones which can be leveraged later in automated build pipelines.
Workload Factory accelerates the creation and management of software version data infrastructure, by leveraging FSx for ONTAP's thin cloning as mentioned previously. It eliminates the need for custom tools or scripts and seamlessly integrates with your CI/CD workflows including a built-in integration with Perforce P4 (Helix Core).
This allows you to more easily speed up development lifecycles, lower cloud spend, and increase developer productivity, making it easier to :
- Automate data cloning for software versions: Workload Factory facilitates the cloning of FSx for ONTAP-based data environments without the need for third-party tools or scripts. Developers can instantly create and access required software versions environments.
- Integrate cloning into CI/CD: Workload Factory seamlessly fits into existing CI/CD workflows by offering a well-documented API, as well as an out-of-the -box integration with Perforce P4 (Helix Core).
- Leverage Built-in administrations: Workload Factory simplifies administrational tasks of each software environment including automatic environment space reclamation, capacity limitations, and access duration.
- Establish workspace data control and audit: With Workload Factory, you gain the ability to effectively manage access policies for project’s data, monitor data usage, and audit user activity.
Get started with Workload Factory here: https://console.workloads.netapp.com
In Conclusion
Amazon FSx for NetApp ONTAP and AWS together provide a comprehensive, scalable, and cost-efficient platform tailored for the unique needs of modern game development. From managing massive game code repositories and accelerating build pipelines to enabling agile developer environments, FSx for ONTAP empowers studios to innovate faster and deliver richer player experiences.
By bridging the DevOps gap between traditional software and game development, FSx for ONTAP helps studios reduce costs, improve resilience, and accelerate time-to-market—ultimately fueling creativity and success in an increasingly competitive industry.
