I am an assigned OPS-DIRdirectorate reviewer for draft-ietf-ipsecme-multi-sa-performance-09.

Summary: Ready with nits

Overall I found the document clear and easy to read.

Editorial nits:

1) Comment relates to definition or the scope of "per-resource". The abstract states "..with the same Traffic Selectors used on different resources, such as CPUs..", however, the rest of the document really is only about multiple CPU use case. I would rephrase the abstract accordingly something along the lines:
"..with the same Traffic Selectors used on different (hardware or software) resources to increase bandwidth of IPSec traffic between peer. This document uses a CPUs as an example of a resources."
And the reword other places accordingly. 

2) the statement in Section 3:
"This identifying data SHOULD be a unique identifier within all the Child SAs with the same TS payloads". In what case the identifier would not be unique within all Child SAs with the same TS payload?

3) Section 1.2
 - Add SAD, PFP to terms defined in RFC4301
 - SADB_ACQUIRE term should also have a reference here similarly to others

4) SAD is never expanded

5) CP, SA and TSr are expanded multiple times