Qiaobin Fu
Qiaobin Fu
Smart NICs can potentially help accelerate Gatekeeper's performance. Example of Smart NICs: [Netronome Agilio](https://www.netronome.com/products/smartnic/overview/). They can be used to [accelerate DPDK-based networking applications](https://www.netronome.com/solutions/dpdk-networking-acceleration/), increasing throughput (Mpps) and bandwidth (Gb/s) to...
When I tried to evaluate the performance of Gatekeeper with table size 2^29, Gatekeeper fails to run with the following error message: `EAL: eal_memalloc_alloc_seg_bulk(): couldn't find suitable memseg_list`. After writing...
In the pull request #89 , when I tried to call ``luaL_get_ctypeid()`` on ``struct gk_config *`` and ``struct gk_fib_dump_entry *``, both return values are the same (i.e., 166) even they...
In GT block, each GT instance needs to keep one Lua state to maintain the most recent Lua policies. Currently, the Lua states are allocated by calling ``luaL_newstate()``. However, this...
Gatekeeper may benefit from some new DPDK features: 1. [Fast mbuf free](https://doc.dpdk.org/guides/nics/features.html#fast-mbuf-free): This feature requires that per queue, all mbufs come from the same mempool and has refcnt = 1....
After Gatekeeper processing the incoming packets, both request packets and granted packets will be sent to Grantors using IP-in-IP encapsulation. In the extreme case, all the packets from a Gatekeeper...
To solve the issue #374 for the first deployment, pull request #390 tries to have multiple SOL blocks to spread the load from GK blocks. Currently, each SOL block enforces...
In #139 using an lcore just for a timer is quite wasteful. For now, we are going to add this burden on the LLS block. This issue is meant to...
In the MICA II paper "Architecting to Achieve a Billion Requests Per Second Throughput on a Single Key-Value Store Server Platform”, one optimization technique they used is to change the...
Currently, we are using a single global [LPM table](https://doc.dpdk.org/guides/prog_guide/lpm_lib.html) in GK blocks, which uses a variation of the DIR-24-8 algorithm. Through the evaluation in the paper titled [DXR: Towards a...