PfSense 3.0 Roadmap



  • I know I'm a bit late to the party and apologies if this has been discussed, I have searched, but I was just reading this blog post https://blog.pfsense.org/?p=1588 and had a couple of questions as it seems pfsense is going from strength to strength.

    • Why the choice of Intel DPDK over something like netmap or dna -> does DPDK perform better? what about the license for DPDK, I thought there was certain things only accessible with purchase of a premium license?

    • I like the mention of moving to a model closer to crochet, would that potentially make it easy to get pfSense on ARM or MIPS64 as well as AMD64?

    • Bootstrap + Python FTW!



  • I know I'm a bit late to the party and apologies if this has been discussed, I have searched, but I was just reading this blog post https://blog.pfsense.org/?p=1588 and had a couple of questions as it seems pfsense is going from strength to strength.

    Why strength? This are really good news for me personally. I would be glad about all functions, options
    and features named in that Blog you linked to! It matches all new Intel based CPUs likes the Intel Xeon D-15xx
    series and the Skylake based Xeons.

    • Better AES-NI support for AES-GCM
    • Using AVX/AVX2 to use the DPDK for a faster Layer2/3 routing part
    • Using the Intel QuickAssist integrated in CPUs or on Intel QuickAssist adapters.
    • Why the choice of Intel DPDK over something like netmap or dna -> does DPDK perform better? what about the license for DPDK, I thought there was certain things only accessible with purchase of a premium license?

    Not named or shown in this Blog, but they are future going or thinking at netmap and DPDK, so
    both will be inserted to the pfSense source code for sure not all in the same time.

    • I like the mention of moving to a model closer to crochet, would that potentially make it easy to get pfSense on ARM or MIPS64 as well as AMD64?

    Could be also interesting, I was also asking many times for this, but it seams that this would be to much
    at this time for them to realize it all.



  • @MAHDTech:

    I know I'm a bit late to the party and apologies if this has been discussed, I have searched, but I was just reading this blog post https://blog.pfsense.org/?p=1588 and had a couple of questions as it seems pfsense is going from strength to strength.

    • Why the choice of Intel DPDK over something like netmap or dna -> does DPDK perform better? what about the license for DPDK, I thought there was certain things only accessible with purchase of a premium license?

    • I like the mention of moving to a model closer to crochet, would that potentially make it easy to get pfSense on ARM or MIPS64 as well as AMD64?

    • Bootstrap + Python FTW!

    I believe your question was answered in a later blog. In https://blog.pfsense.org/?p=1866 it seems like pfsense 3.x is leaning more towards netmap or even a combination of both.

    Back in February, I wrote a blog post that discussed our plans for pfSense software version 2.3, which is now in alpha, and our plans for pfSense 3.0.  While I promoted DPDK then, we’ve since found that netmap provides a simpler API, and substantially better safety, as the device drivers remain in the kernel, rather than running in userspace with DPDK.  Still, DPDK provides a set of libraries, such as longest-prefix match, which uses a variation of the DIR-24-8 algorithm for routing lookups, which we should find useful in our pursuit of the ultimate open source software router.

    Carlos


Log in to reply