SNIA Developer Conference September 15-17, 2025 | Santa Clara, CA
SMB3 has seen significant adoption as the storage protocol of choice for running private cloud deployments. With the recent advances in persistent memory technologies, we will take a look at how we can leverage the SMB3 protocol in conjunction with SMBDirect/RDMA to provide very low latency access to persistent memory devices across the network. With the increasing popularity of cloud storage - technologies like Azure Files which provide seamless access to cloud stored data via the standard SMB3 protocol is seeing significant interest. One of the key requirements in this space is to be able to run SMB3 over a secure / firewall friendly internet protocol. We will update you on some work we are doing to enable SMB3 over QUIC - which is a recent UDP based transport with strong security and interop properties. We will explore some work we have done to enable on-the-wire compression for SMBDIRECT SMB3. Learning Objectives: 1. Learn how we can use SMB3 to setup direct RDMA access to remote persistent memory 2. Using QUIC as a transport for SMB3 3. How can we use data compression algorithms to optimize SMB data transfer?
SMB3 has seen significant adoption as the storage protocol of choice for running private cloud deployments. With the recent advances in persistent memory technologies, we will take a look at how we can leverage the SMB3 protocol in conjunction with SMBDirect/RDMA to provide very low latency access to persistent memory devices across the network. With the increasing popularity of cloud storage - technologies like Azure Files which provide seamless access to cloud stored data via the standard SMB3 protocol is seeing significant interest. One of the key requirements in this space is to be able to run SMB3 over a secure / firewall friendly internet protocol. We will update you on some work we are doing to enable SMB3 over QUIC - which is a recent UDP based transport with strong security and interop properties. We will explore some work we have done to enable on-the-wire compression for SMBDIRECT SMB3. Learning Objectives: 1. Learn how we can use SMB3 to setup direct RDMA access to remote persistent memory 2. Using QUIC as a transport for SMB3 3. How can we use data compression algorithms to optimize SMB data transfer?
Starting with version 4.14 Samba provides core infrastructure code that allows basing all access to the server's filesystem on file handles and not on paths. An example of this is using fstat() instead of stat(), or SMB_VFS_FSTAT() instead of SMB_VFS_STAT() in Samba parlance. Historically Samba's fileserver code had to deal a lot with processing path based SMB requests. While the SMB protocol itself has been streamlined to be purely handle based starting with SMB2, large parts of infrastructure code remains in place that will "degrade" handle based SMB2 requests to path based filesystem access. In order to fully leverage the handle based nature of the SMB2 protocol we came up with a straight forward way to convert this infrastructure code, so it can be converted to make use of a purely handle based VFS interface. The talk will present what we have achieved so far and what is left to do. It's intented audience is anyone working on the Samba fileserver code and anyone working on Samba VFS modules.
Over the past year many improvements have been made in Linux for accessing files remotely. This has been a great year for cifs.ko with the addition of new SMB3.1.1 features and optimizations. It continues to be the most active network/cluster file system on Linux. Improvements to performance with handle leases (deferred close), multichannel, signing improvements, huge gains in read ahead performance, and directory and metadata caching improvements have been made. And security has improved with support for the strongest encryption, and more recently the exciting work on QUIC. Many other security improvements have been added and will be described. This presentation will go through the features added over the past year to the Linux client, and demonstrate how they help common scenarios, from accessing the cloud faster (like Azure) to accessing Samba, Windows, Macs and the new Linux kernel server (ksmbd).
Samba had experimental support for multi-channel for quite a while. SMB3 has a few concepts to replay requests safely. We now implement them completely (and in parts better than a Windows Server). The talk will explain how we implemented the missing features. With the increasing amount of network throughput, we'll reach a point where a data copies are too much for a single cpu core to handle. This talk gives an overview about how the io_uring infrastructure of the Linux kernel could be used in order to avoid copying data, as well as spreading the load between cpu cores. A prototype for this exists and shows excellent results.
SMB3 has seen significant adoption as the storage protocol of choice for running private cloud deployments. In this iteration of the talk, we’ll update the audience on SMB protocol changes as well as improvements to the Windows implementation of the SMB server and client. Added to the SMB protocol is a new server-to-client notification mechanism, which enables a variety of novel use cases. We’ll present the details of protocol messaging (new message types, etc) as well as the one scenario which leverages this new mechanism (server-triggered graceful session closure). In addition to the protocol changes, we’ll provide an overview of the latest feature additions to the Windows SMB server and client: authentication rate limiting to protect against password spray attacks, and upcoming improvements to SMB over QUIC.
One fallout of Samba's protection against symlink races was a slowdown of pathname handling. This talk will present the current status at the time of the conference of the ongoing work to fix the slowdown by using modern Posix and in particular Linux/FreeBSD APIs.