Estou tendo problemas de lentidão usando nginx + php-fpm. Com cerca de 20 mil acessos simultâneos o site fica lento, bastante lento, sei que isto é devido aos processos do php-fpm. O código php é limpo e bem estruturado, rodo o site em uma máquina com as seguintes configurações: 8GB de ram, 6vcpus, rede 1gbps. 

Configuração do meu php-fpm: 

; The number of child processes to be created when pm is set to 'static' and the
; maximum number of child processes when pm is set to 'dynamic' or 'ondemand'.
; This value sets the limit on the number of simultaneous requests that will be
; served. Equivalent to the ApacheMaxClients directive with mpm_prefork.
; Equivalent to the PHP_FCGI_CHILDREN environment variable in the original PHP
; CGI. The below defaults are based on a server without much resources. Don't
; forget to tweak pm.* to fit your needs.
; Note: Used when pm is set to 'static', 'dynamic' or 'ondemand'
; Note: This value is mandatory.
pm.max_children = 75

; The number of child processes created on startup.
; Note: Used only when pm is set to 'dynamic'
; Default Value: min_spare_servers + (max_spare_servers - min_spare_servers) / 2
pm.start_servers = 20

; The desired minimum number of idle server processes.
; Note: Used only when pm is set to 'dynamic'
; Note: Mandatory when pm is set to 'dynamic'
pm.min_spare_servers = 20

; The desired maximum number of idle server processes.
; Note: Used only when pm is set to 'dynamic'
; Note: Mandatory when pm is set to 'dynamic'
pm.max_spare_servers = 35

; The number of seconds after which an idle process will be killed.
; Note: Used only when pm is set to 'ondemand'
; Default Value: 10s
pm.process_idle_timeout = 5s;

; The number of requests each child process should execute before respawning.
; This can be useful to work around memory leaks in 3rd party libraries. For
; endless request processing specify '0'. Equivalent to PHP_FCGI_MAX_REQUESTS.
; Default Value: 0
pm.max_requests = 500

rlimit_files = 350000
rlimit_core = unlimited

Também uso o opcache, fiz alterações no kernel, nginx etc... Tudo configurado conforme os recursos da máquina. Quando olho o consumo não fica nem 40% de cpu/ram/rede.

O Nginx faz cache? Se não o faz, considere isso.