SeaStar is an event-driven framework allowing you to write non-blocking, asynchronous code in a relatively straightforward manner (once understood). It is based on futures.
See instructions for Fedora, CentOS and Ubuntu.
There are also instructions for building on any host that supports Docker.
Seastar can be built with the C++17 dialect by supporting compilers, conditional
on the --c++-dialect option being set to gnu++17.
However, by default Seastar uses C++14-compatible types such as
std::experimental::optional<> or boost::variant, both internally and in its public
API, thus forcing them on C++17 projects. To fix this, Seastar provides the --use-std-optional-variant-stringview 0|1
option, which changes those types to their stdlib incarnation, and allows
seemless use of C++17. Usage of this option requires an updated compiler, such
as GCC 8.1.1-5 on Fedora.
There is a mini tutorial and a more comprehensive one.
The documentation is available on the web.
Ask questions and post patches on the development mailing list. Subscription information and archives are available here, or just send an email to seastar-dev@googlegroups.com.
Information can be found on the main project website.
File bug reports on the project issue tracker.
Seastar comes with its own userspace TCP/IP stack for better performance.
- CPUs - As much as you need. SeaStar is highly friendly for multi-core and NUMA
- NICs - As fast as possible, we recommend 10G or 40G cards. It's possible to use 1G too but you may be limited by their capacity. In addition, the more hardware queue per cpu the better for SeaStar. Otherwise we have to emulate that in software.
- Disks - Fast SSDs with high number of IOPS.
- Client machines - Usually a single client machine can't load our servers. Both memaslap (memcached) and WRK (httpd) cannot over load their matching server counter parts. We recommend running the client on different machine than the servers and use several of them.
On Travis CI:
