Often, the most impactful platform updates are the ones that remove complexity while expanding what's possible. Harper’s 4.5 release is a perfect example. It doesn’t just tack on blob storage as a checkbox feature—it simplifies the handling of unstructured content at scale while maintaining Harper’s signature performance and ease of development.
I sat down with Kris Zyp, SVP of Engineering at Harper, to discuss the “why” and “how” behind this release. What emerged was a powerful narrative about the evolving needs of the web and how Harper is positioning itself as the fused engine behind the next generation of performant, content-heavy applications.
Why Blob, Why Now?
Harper has long been known for its high-performance and fused data and application for modern web architectures. But as Kris pointed out, structured data don’t just power today’s web apps—they’re also rich with media.
“We’ve been powering more significant websites and e-commerce sites,” Kris explained. “That space requires not just structured data like product info and user profiles, but also a lot of unstructured data—images, videos, and even rendered HTML. Blob Storage unlocks the ability to handle that efficiently.”
In other words, Harper is meeting developers where they are—at the intersection of structured and unstructured content. And it's doing so with a native blob storage system that’s designed not as a bolt-on but as a deeply integrated extension of the Harper stack, taking advantage of the streaming, sharding, and replication functionality already part of Harper.
Streaming as a Paradigm Shift
One of the more exciting aspects of Harper’s approach is how it embraces a streaming-first architecture. Traditional databases tend to focus only on atomic operations—great for deterministic data manipulation but ill-suited for large files or real-time media.
“When you're dealing with larger chunks of data—especially video—streaming becomes a much more natural model,” Kris noted. “You can upload a video, record the metadata, and start streaming it to clients before the upload is even complete.”
This opens the door to real-time content delivery workflows, like live video ingest and playback, without the need to build and maintain a sprawling set of edge services and CDNs. Harper is simplifying the architecture, minimizing the development and infrastructure overhead that tends to slow down growth and balloon costs.
Built for Web Performance
E-commerce and high-traffic websites are particularly sensitive to latency and load times, and Harper's Blob Storage is optimized to meet those demands.
“We’re seeing great results caching full HTML pages as blobs,” Kris shared. “Some of these pages are a few hundred kilobytes, and we’ve been able to update thousands per second—basically hitting network transfer limits.”
That’s no small feat. On 16-core, 32GB machines using Akamai Connected Cloud's block storage, Harper achieved performance that maxed out available bandwidth, not CPU or disk. In other words, the system is highly efficient—something any performance-minded dev or SRE will appreciate.
And because Harper is inherently a distributed system, that performance scales horizontally.
“You can shard your data, replicate blobs in real time, and serve users from the closest geographic node,” Kris said. “It all ties into our goal of delivering a low-latency experience globally.”
Real Use Cases: E-Commerce, Caching, Media Delivery
Let’s talk about practical applications. Harper’s blob storage shines in:
- E-commerce: Hosting product images, videos, and HTML caches of frequently visited pages, reducing compute costs and improving load times.
- Media platforms: Uploading, storing, and streaming large media files, including real-time video.
- Dynamic page generation: Caching rendered HTML from frameworks like Next.js for ultra-fast page delivery.
- Global content delivery: Harper’s distributed nodes ensure fast, local access to blob content.
All of this is powered by a concurrency-focused storage engine with deep queuing and high I/O throughput—optimized not just for performance but for scale.
What Developers Should Know
Of course, working with large files introduces new paradigms. Kris offered a word of advice:
“(When streaming blobs), you're operating outside the typical ACID model. Uploads can be interrupted. Harper provides good facilities for reporting those errors, but developers should plan for those edge cases—like deciding what to do with truncated uploads.”
Harper gives you the tools to handle these conditions gracefully, whether that means retry logic, cleanup processes, or partial data strategies.
What Else, What Next? The Road to 4.6 and Beyond
On top of it all, observability has also taken a leap forward in 4.5. Harper now adaptively tracks storage usage and can evict stale content from cache to free up space—automatically.
“It reduces the operational overhead,” Kris explained. “You can be more aggressive with caching because Harper will manage capacity for you.”
Further, Blob Storage isn’t the end—it’s part of a much larger vision to streamline development across use cases. Kris hinted at what’s coming next:
“We’re planning further integrations for web frameworks, easier HTML and content caching, and more out-of-the-box functionality to accelerate web app development.”
We continue to move toward a future where Harper doesn’t just power data—it powers your entire web application, front to back, structured to unstructured, atomic to streamed.
Final Thoughts
Harper 4.5 is more than a dot-release—it’s another significant leap toward a more unified, performant, developer-friendly web stack. Blob Storage is not just a new feature; it’s a strategic capability that extends Harper’s vision of a fused data and content platform.
Whether you're building a product catalog, launching a video platform, or optimizing a blazing-fast storefront, Harper’s Blob Storage has you covered—streaming-first, distributed by default, and built to scale. Explore Harper today!
.jpg)
.png){kind=icon}
.png){kind=icon}