Our products

Achieve Universal Data Movement

products overview

Radically Simple Data Mover That Does More

Enterprises, national labs, supercomputing centers depend on Zettar to move their data at speed and scale with automation, unleashing operational efficiency. We build data mover software solutions that are radically simple, cloud-ready, and provide API-enabled automation.

Cloud-Readiness

Our solutions are designed with cloud in mind right from the beginning. Use the cloud, including private cloud and hybrid cloud efficiently and economically. They work with compute and storage clouds, on-prem, hybrid, and public,

API-enabled Automation

The unique peer-to-peer architecture enables a dramatically simple RESTful API-enabled approach. Each zx instance has a built-in Web server and any zx instance in a cluster can be the management entry point - operating a cluster made easy.

Data Mobility at Performance

We enable moving data transparently at speed and scale between file and cloud storage, on prem, private, hybrid, public. All are covered. Movements are always carried out at the maximum performance possible. Workflows are accelerated, always.

All in One

Zettar has all four products, zx-File, zx-Object, zx-Single-site-copy, and zx-Streaming integrated into one, namely zx. Note that the functionality of the last three high-value, logically separated products must be enabled with appropriate licenses.

products
zx-File supports any POSIX compliant file services, e.g. XFS, Lustre, NFSv3/4. It bulk transfers or streams data over digital connections of any speed accross any distance. It is the only commercial data mover that has transferred actual 1PB over 5000-miles in 29 hours at 94% bandwidth utilization (under a 80Gbps bandwidth cap) with full TLS encryption and checksumming. It can replicate data sets incrementally, even those with hundreds of millions of small files.
zx-Object supports AWS S3 object storage and true compatibles. It transfers data objects over digital connections of any speed across any distance. It supports the advanced AWS S3 multipart REST APIs highly concurrently resulting both very high reading and writing peformance. It is the only product of this kind that has been proven work well with Samsung’s KV SSDs, with the results demonstrated at the Supercomputing 2019 (SC19) Conference.

zx has highly efficient and rich ways to interact with storage services. The embedded software in today’s physical data shipping devices that we know of has poor efficiency in this regard. Thus, both the data loading at one end and data offloading at the other end are time consuming and tedious. With zx-single-site copy, it accelerates both operations for such data shipping devices (AWS Snowball alike).

Recent advances in all aspects of computing have made it possible to stream modern machine generated data from multiple sources through various stages of online and near line processing. This results in rich multi-dimensional datasets that can be made accessible for processing using a hybrid cloud. zx-Streaming facilitates such a modern workflow model.
Strength

zx is the most modern scale-out petascale-proven data mover software available in the market.  Its peer-to-peer enabled scale-out architecture, robustness, performance, simplicity, and capabilities are proven with production deployments, many world records, and winning of international competitions.

Group 2Created with Sketch.

Simple to Use

Once setup and configured, to users of zx, either via its built-in Web UI or its RESTful APIs, the effort is much less than using typical CLI data mover tools or other competing products owing to zx’s unique peer-to-peer enabled scale-out architecture. The recent releases support both file and object storage in a simple and consistent manner. Please Request demo and see for yourself how easy it is to use zx. It also shows the intuitive built-in Web UI.

Highly Efficient

Assuming sufficient storage throughput available, a decent CPU model, fully populated CPU memory channels, enough available network bandwidth, and normal file/object sizes, a single data transfer node running zx can comfortably push/pull 70+Gbps or beyond. If two nodes are employed at each end, such a set up not only has the potential to go beyond 100Gbps but also provides high-availability (HA) at the same time. Such high-efficiency translates into high ROI.

Peer-to-Peer Enabled Scale-Out

zx is one of the only three scale-out data mover software applications (the most modern one) that have been funded by the U.S. DOE Office of Science. Owing to its unique peer-to-peer architecture, running a cluster is as simple as running a single instance - the multiple zx instances are self-organizing and self-healing. Rich run-time statistics and connection graphs are all included. The architecture also provides the highly desirable linear scalability.

Get started now

No Fear to Data Gravity