Accessiblity: ObjectiveFS is available in all regions and supports cross-region access. You can access your data from anywhere (any data center region, your office and your laptop). Amazon EFS is only available in 5 regions and can only be accessed by EC2 instances in the same region.
Performance: ObjectiveFS has high performance that scales. It is 80X faster for reads and 69X faster for writes for web servers, CMS and small file workloads compared to EFS (see details). For large files workloads such as videos, genomic data, ObjectiveFS has over 350MB/s throughput and can sustain high throughput 24x7 (see details). EFS performance is limited by burst limit based on the filesystem size (reference: Amazon EFS performance).
Storage Cost: ObjectiveFS doesn’t charge for storage, and you only pay the Amazon S3 storage cost (~$0.03/GB). Amazon EFS charges ~$0.30/GB.
Durability: ObjectiveFS uses highly durable object stores like Amazon S3 with 99.999999999% durability for storage. Amazon EFS is NFS-based.
Others: ObjectiveFS has been running in production since 2013. It also supports other S3-compatible object stores such as Cloudian, Google Cloud Storage, IBM Cleversafe, Ceph, etc.
ObjectiveFS is a POSIX-compatible file system that provides a standard file system interface. It is a log-structured filesystem on top of Amazon S3 that we implemented from scratch, and works with Linux and macOS programs and tools.
S3FS is a way of viewing your S3 bucket as a file system to make it easier to manage the objects in your bucket. While S3FS has a large subset of POSIX support, it is a one-to-one mapping, and doesn’t fully support regular file system semantics or consistency guarantees (e.g. atomic rename of directories, mutual exclusion of open exclusive, append to file requires rewriting the whole file and no hard links). This breaks the expectations of many Linux programs.
ObjectiveFS has built-in strong encryption and end-to-end data integrity check that are on by default.
ObjectiveFS supports concurrent mounts and read/write from multiple machines. S3QL supports only 1 node.
ObjectiveFS also keeps the file system data consistent and stores it in S3 as soon as possible, for sharing and to handle EC2 unexpected termination. S3QL uploads all metadata changes at a regular interval (by default every 24h) or during unmount.
ObjectiveFS simplifies cluster management by offloading the storage cluster to the cloud providers. With GlusterFS and CephFS, you need to maintain your own storage cluster and need extra machines to store your data.
ObjectiveFS is a live filesystem, where any update is visible to all sharers. We also provide standard semantics (POSIX) to handle file operations and provide consistency guarantees. So, standard tools and software can run directly on it.
Box and Dropbox are file-sharing focused, and use the syncing filesystem approach. In a syncing filesystem, updates are synchronized in the background and potential conflicts may occur.
With ObjectiveFS, you can easily share file assets among your servers. Just mount ObjectiveFS on your servers, and they can all read and write the same file assets.
ObjectiveFS also automatically scales with your file asset storage, so you don’t need to worry about running out of disk space.
Yes. You can easily share your WordPress/Drupal/Joomla/Magento file assets between servers with ObjectiveFS. See how to scale CMS with ObjectiveFS.
We have 3 plans: Startup, Business and Enterprise.
The Startup plan starts at $29 and includes 2 instances. The Business plan starts at $299 and includes 10 instances. Both plans have volume discounts for instances over 25. For details, please see the pricing page.
We also have a cluster plan if you have dynamic workloads and plan to spin up many EC2 instances for temporary jobs. Please email firstname.lastname@example.org if you would like to learn more.
We only count the number of active ObjectiveFS copies that you run concurrently. If you run two copies of ObjectiveFS at different times and they do not overlap, it is counted only as 1 instance.
Instances can be of the same or different file systems. For example, 3 servers all running the same file system, or each running a different file system both count as 3 instances.
Only the 99-percentile of the max instances is counted. So short usage spikes (around 7 hours per month) are filtered out.
S3 and GCS charges are billed separately by AWS and Google, and are not included in our pricing.
Yes. We have a 14-day free trial. No credit card is required. You can sign up here.
You get unlimited instances and support, so you can fully evaluate ObjectiveFS.
If you no longer wish to use ObjectiveFS, you can cancel at any time.
You are never locked in to ObjectiveFS. You can always copy your data back to your local storage or other file systems using your favorite tools: tar, rsync, cp, pax, cpio, etc.
Yes, you can always copy your data back to your local storage or other file systems using standard tools (e.g. tar, rsync, etc).
Our company has been in business for over 13 years and we are in it for the long haul.
In the extremely unlikely event that we cease operation, all active users can continue using the software and keep accessing their data using ObjectiveFS.
No, your license will remain the same. When you subscribe to a plan, our system automatically upgrades your license. So, there is no need to change your setup and everything will continue to work.
Yes, the filesystems that you created during your trial will continue to work. Uptime for our users is very important to us. So, everything you’ve set up during your trial will continue to work when you subscribe.
Your data will be safely stored in your S3/GCS bucket. You can always copy your data back to your local storage or other file systems using regular tools (e.g. tar, rsync, cp, etc) from ObjectiveFS if you decide not to subscribe after the trial.
If you decide to use ObjectiveFS in the future, you can still access the data you have created during the trial if your S3/GCS bucket hasn’t changed.
ObjectiveFS is a log-structured filesystem and uses an S3 bucket as a storage backend. To provide POSIX compatibility and to make sure data is not corrupted during transit, all data and metadata are compressed, encrypted and checked by our end-to-end data integrity checker.
ObjectiveFS expects to manage all contents of the S3 bucket that it is using for storage, and to have only its files in that S3 bucket.
No. ObjectiveFS is different from bucket viewers like S3FS and uses the S3 bucket as a storage backend (see description above for details).
No. ObjectiveFS is a log-structured filesystem and uses S3 buckets as a storage backend (see details). So, copying files directly into the S3 bucket will not work with this design.
No. Since ObjectiveFS is a log-structured filesystem, it expects to manage all contents of the bucket and to have only its files in the bucket. Please store non-ObjectiveFS files in a different S3 bucket.
No. ObjectiveFS uses S3 as a storage backend and implements a log-structured backend to provide features like snapshots. It also can store many small files in 1 object. Therefore, there is not a 1-to-1 mapping between a file and an S3 object in your S3 bucket.
ObjectiveFS implements strong checksums to detect data corruption and to verify that the data retrieved from S3 matches with the originally stored data. This makes sure that the data integrity remains intact. ObjectiveFS also verifies that the first level index matches with their checksums when a filesystem is freshly mounted. Any checksum mismatches will always be reported in the log and the operation will be retried.
If your S3 bucket is in a region affected by an S3 outage, ObjectiveFS will always retry requests to S3. So, any pending reads and writes to S3 will be retried until S3 is back up again as long as the filesystem remains mounted on your server.
One useful tip: you can set the
AWS_DEFAULT_REGION environment variable to the same region as your S3 bucket in ObjectiveFS 5.0 or newer. This avoids querying the S3 global endpoint (s3.amazonaws.com) at filesystem start up, which can reduce the impact of an outage if your region is still fully functional.
No. Your data is encrypted locally on your server by ObjectiveFS. The encrypted data is sent directly to AWS S3 or GCS and stays encrypted until it returns to one of your servers.
ObjectiveFS uses the NaCl crypto library, with modern algorithms such as Salsa20 and Poly1305, to keep your data secure. This approach has no data-dependent branches or data-dependent array indices, and protects against cache timing attacks that some AES implementation is vulnerable to.
ObjectiveFS implements strong checksums to detect data corruption and to verify that the data retrieved from S3 matches with the originally stored data. This makes sure that the data integrity remains intact.
A successful mount of an ObjectiveFS is similar to a successful fsck. When a filesystem is mounted, ObjectiveFS automatically verifies the checksums of the first level index. When data is read from your filesystem (on either a new mount or an existing mount), ObjectiveFS also verifies the checksums to detect any data corruption. Any checksum mismatches will always be reported in the log and the operation will be retried.
Very fast. The performance is similar to accessing a local disk. We use workload adaptive heuristics, advanced caching and bundled writes to achieve this performance.
Currently no. However, you can run ObjectiveFS on a linux machine and export to Windows via NFS (see details) or Samba (see details). Please email email@example.com if you would like native Windows support.
If you have a question, please email us at firstname.lastname@example.org.