Enterprise

MikroTik CCR2116-12G-4S+

Roteador carrier-class com 16 cores ARM, L3 Hardware Offloading e 51 Gbps de throughput

MikroTik CCR2116-12G-4S+
Classe enterprise para ISPs medios e grandes. O CCR2116-12G-4S+ entrega 51 Gbps com L3 Hardware Offloading, BGP 6x mais rapido que antecessores e fonte redundante. Projetado para ser o core router da sua rede.

O Que Vem na Caixa

Especificacoes Tecnicas

Especificacao CCR2116-12G-4S+
CPU AL73400 (Amazon Annapurna Labs Alpine), ARM 64-bit, 16 cores, 2000 MHz
RAM 16 GB DDR4
Storage 128 MB NAND
Switch Chip Marvell 98DX3255 (Aldrin)
Portas Ethernet 13x Gigabit: 12 via switch chip (3 grupos de 4) + 1 direta ao CPU (gerencia)
SFP+ 4x SFP+ 10G (cada uma full-duplex direto ao switch chip)
USB Nenhum (modelo padrao)
M.2 1x PCIe Gen 3.0 x4 (NVMe 2280)
Serial Console RJ45
Alimentacao 2x AC 100-240V (dual redundante com failover automatico)
Consumo maximo 83W (60W tipico)
Form Factor Rackmount 1U, 443 x 199 x 44 mm, 4 ventiladores, peso 3.2 kg
Temperatura -20 C a +60 C
RouterOS Licenca Level 6 (somente v7)
Recursos Especiais L3 Hardware Offloading (51 Gbps), M.2 NVMe, BGP 6x mais rapido, fonte dual

Alimentacao e Redundancia

Entrada Tensao Observacao
Fonte AC 1 (Principal) 100-240V AC Conector IEC padrao, hot-swap
Fonte AC 2 (Redundante) 100-240V AC Failover automatico em caso de falha
Consumo Tipico 60W Em operacao normal
Consumo Maximo 83W Carga total com 4 SFP+ ativas
Redundancia de energia real. As duas fontes AC operam simultaneamente. Se uma fonte falhar, a segunda assume instantaneamente sem queda de servico. Para ambientes criticos de ISP, isso elimina a necessidade de fontes externas redundantes.

Diagrama de Bloco

AL73400 (Annapurna Labs Alpine) 16 cores ARM64 | 2000 MHz CPU Principal RAM 16 GB DDR4 Memoria Principal NAND 128 MB Armazenamento M.2 NVMe (PCIe 3.0 x4) Slot 2280 High-Speed Bus Marvell 98DX3255 (Aldrin) Switch Chip - L3 Hardware Offloading 51 Gbps throughput offloaded Ether1 (Gerencia) 1 GbE direto ao CPU Grupo 1: ether2 - ether5 4x Gigabit (via switch chip) Grupo 2: ether6 - ether9 4x Gigabit (via switch chip) Grupo 3: ether10 - ether13 4x Gigabit (via switch chip) 4x SFP+ 10G (sfp-sfpplus1 a sfp-sfpplus4) Cada porta full-duplex direto ao switch chip Serial RJ45 Alimentacao - Dual AC Redundante com Failover AC 1: 100-240V (Principal) AC 2: 100-240V (Redundante)

Benchmarks Ethernet

Fonte: MikroTik (testes em laboratorio com RouterOS v7). Valores em Mbps.

Metodologia: testes bidirecionais com multiplos fluxos, hardware em temperatura ambiente controlada (25 C). Pacotes de tamanho fixo conforme coluna.

Modo Configuracao 1518B (Mbps) 512B (Mbps) 64B (Mbps)
Bridging none (fast path) 38.919,1 37.333,4 16.025,6
Bridging 25 bridge filter rules 32.333,4 10.986,3 1.500,8
Routing none (fast path) 39.009,0 37.377,2 15.284,0
Routing 25 simple queues 39.009,0 25.897,4 3.427,7
Routing 25 ip filter rules 39.009,0 16.088,7 2.122,4
Routing none (L3HW Offloading) 51.323,7 50.045,1 40.181,8

Visualizacao comparativa - Routing 1518B (Mbps):

Fast Path (1518B) 39.009 Mbps
25 Simple Queues (1518B) 39.009 Mbps
25 IP Filter Rules (1518B) 39.009 Mbps
L3 Hardware Offloading (1518B) 51.323 Mbps
L3 Hardware Offloading: o diferencial. O chip Marvell Aldrin faz roteamento L3 inteiramente em hardware, atingindo 51,3 Gbps com pacotes de 1518 bytes e impressionantes 40,1 Gbps mesmo com pacotes minimos de 64 bytes. Isso libera os 16 cores do CPU para processamento de regras de firewall, queues e BGP.

Benchmarks IPsec

Fonte: MikroTik (testes em laboratorio com RouterOS v7). Valores em Mbps.

Metodologia: tuneis IPsec site-to-site com criptografia simetrica, temperatura controlada (25 C).

Configuracao Criptografia 1400B (Mbps) 512B (Mbps) 64B (Mbps)
1 tunel AES-128-CBC + SHA1 2.012,6 763,5 93,6
256 tuneis AES-128-CBC + SHA1 4.108,2 2.733,3 342,4
256 tuneis AES-256-CBC + SHA256 4.101,4 2.744,3 343,2

Escala: 0 - 5.000 Mbps | Visualizacao IPsec 256 tuneis (1400B):

256 tuneis AES-128 (1400B) 4.108 Mbps
256 tuneis AES-256 (1400B) 4.101 Mbps
1 tunel AES-128 (1400B) 2.012 Mbps
VPN concentrator de alta performance. Os 16 cores escalam linearmente: com 256 tuneis simultaneos, o throughput chega a 4,1 Gbps - ideal para concentrar VPNs de filiais e POPs. A diferenca entre AES-128 e AES-256 e praticamente zero, entao use AES-256 sem penalidade.

Sugestao de Aplicacao

Core Router de ISP
Com 16 cores, 16 GB RAM e 4 portas SFP+ 10G, o CCR2116 e o roteador ideal para o nucleo de ISPs medios e grandes. A licenca Level 6 libera todos os recursos do RouterOS.
BGP Full Table
Processamento BGP 6x mais rapido que as geracoes anteriores (CCR1xxx). Os 16 GB de RAM comportam facilmente a tabela BGP full + rotas de clientes, mesmo com projecao de crescimento.
Concentrador PPPoE
Capacidade para 500+ clientes PPPoE simultaneos. Os 16 cores distribuem o processamento de autenticacao e encapsulamento, enquanto o fast path acelera o forwarding.
Firewall Enterprise
Com L3 Hardware Offloading, o trafego roteado atinge 51 Gbps direto no switch chip. Os 16 cores ficam dedicados ao processamento de regras de firewall, queues e mangle.
VPN Concentrator
4,1 Gbps com 256 tuneis IPsec simultaneos. Ideal para concentrar VPNs de filiais, POPs remotos e parceiros de peering, com AES-256 sem penalidade de performance.
Backbone com NVMe
O slot M.2 NVMe permite instalar SSD para logs extensivos, traffic flow, capturas de pacotes e armazenamento local de graficos. Ideal para POPs que precisam de diagnostico avancado.

Scripts de Configuracao RouterOS v7

Copie e cole diretamente no terminal do RouterOS. Todos os scripts foram validados para RouterOS v7.x no CCR2116-12G-4S+.

a) Configuracao Basica
/system identity set name="CCR2116-CORE"
/system clock set time-zone-name=America/Sao_Paulo
/system ntp client set enabled=yes
/system ntp client servers add address=a.ntp.br
/system ntp client servers add address=b.ntp.br

# Interface de gerencia dedicada (ether1 direto ao CPU)
/ip address add address=10.255.255.1/30 interface=ether1 \
    comment="Gerencia - direto ao CPU"
/interface set ether1 comment="MANAGEMENT"
b) DNS
/ip dns set servers=8.8.8.8,8.8.4.4,1.1.1.1 \
    allow-remote-requests=yes \
    cache-size=16384KiB \
    cache-max-ttl=1d
c) Rota Default (Principal + Backup)
# Rota principal via SFP+ 1
/ip route add dst-address=0.0.0.0/0 gateway=<GATEWAY-PRINCIPAL> \
    distance=1 comment="Default - Link Principal"

# Rota backup via SFP+ 2
/ip route add dst-address=0.0.0.0/0 gateway=<GATEWAY-BACKUP> \
    distance=2 comment="Default - Link Backup"

# Check gateway para failover automatico
/ip route set [find comment="Default - Link Principal"] \
    check-gateway=ping
/ip route set [find comment="Default - Link Backup"] \
    check-gateway=ping
d) Servidor PPPoE (Multiplos Planos)
# Pools por faixa de plano
/ip pool add name=pool-100m ranges=100.64.1.2-100.64.1.254
/ip pool add name=pool-200m ranges=100.64.2.2-100.64.2.254
/ip pool add name=pool-500m ranges=100.64.3.2-100.64.3.254
/ip pool add name=pool-1g ranges=100.64.4.2-100.64.4.254

# Profiles por plano
/ppp profile add name=plano-100m local-address=100.64.1.1 \
    remote-address=pool-100m dns-server=8.8.8.8,8.8.4.4 \
    rate-limit=100M/100M change-tcp-mss=yes \
    use-compression=default use-encryption=default

/ppp profile add name=plano-200m local-address=100.64.2.1 \
    remote-address=pool-200m dns-server=8.8.8.8,8.8.4.4 \
    rate-limit=200M/200M change-tcp-mss=yes \
    use-compression=default use-encryption=default

/ppp profile add name=plano-500m local-address=100.64.3.1 \
    remote-address=pool-500m dns-server=8.8.8.8,8.8.4.4 \
    rate-limit=500M/500M change-tcp-mss=yes \
    use-compression=default use-encryption=default

/ppp profile add name=plano-1g local-address=100.64.4.1 \
    remote-address=pool-1g dns-server=8.8.8.8,8.8.4.4 \
    rate-limit=1G/1G change-tcp-mss=yes \
    use-compression=default use-encryption=default

# Servidor PPPoE na bridge de clientes
/interface pppoe-server server add service-name=MevloxPPPoE \
    interface=bridge-clientes default-profile=plano-100m \
    authentication=chap,mschap2 max-mtu=1480 max-mru=1480 \
    keepalive-timeout=30 disabled=no

# Exemplos de clientes
/ppp secret add name=cliente001 password=SenhaSegura123! \
    profile=plano-100m service=pppoe
/ppp secret add name=cliente002 password=SenhaSegura456! \
    profile=plano-500m service=pppoe
e) Firewall Padrao Completo
# ===== Interface Lists =====
/interface list add name=WAN
/interface list add name=LAN
/interface list add name=MANAGEMENT
/interface list member add interface=sfp-sfpplus1 list=WAN
/interface list member add interface=sfp-sfpplus2 list=WAN
/interface list member add interface=ether1 list=MANAGEMENT

# ===== Address List - Bogons =====
/ip firewall address-list
add address=0.0.0.0/8 list=bogons comment="RFC1122"
add address=10.0.0.0/8 list=bogons comment="RFC1918"
add address=100.64.0.0/10 list=bogons comment="RFC6598 CGNAT"
add address=127.0.0.0/8 list=bogons comment="Loopback"
add address=169.254.0.0/16 list=bogons comment="Link-Local"
add address=172.16.0.0/12 list=bogons comment="RFC1918"
add address=192.0.0.0/24 list=bogons comment="RFC6890"
add address=192.0.2.0/24 list=bogons comment="TEST-NET-1"
add address=192.168.0.0/16 list=bogons comment="RFC1918"
add address=198.18.0.0/15 list=bogons comment="Benchmarking"
add address=198.51.100.0/24 list=bogons comment="TEST-NET-2"
add address=203.0.113.0/24 list=bogons comment="TEST-NET-3"
add address=224.0.0.0/4 list=bogons comment="Multicast"
add address=240.0.0.0/4 list=bogons comment="Reservado"

# ===== Filter - Input =====
/ip firewall filter
add chain=input connection-state=established,related action=accept \
    comment="Aceita estabelecidas"
add chain=input connection-state=invalid action=drop \
    comment="Descarta invalidas"
add chain=input protocol=icmp limit=50,5:packet action=accept \
    comment="ICMP limitado"
add chain=input protocol=icmp action=drop \
    comment="Excesso ICMP"
add chain=input in-interface-list=MANAGEMENT action=accept \
    comment="Aceita management (ether1)"
add chain=input in-interface-list=LAN action=accept \
    comment="Aceita LAN"
add chain=input src-address-list=bogons in-interface-list=WAN \
    action=drop comment="Drop bogons na WAN"
add chain=input action=drop \
    comment="Descarta resto"

# ===== Filter - Forward =====
/ip firewall filter
add chain=forward connection-state=established,related \
    action=fasttrack-connection comment="FastTrack"
add chain=forward connection-state=established,related action=accept
add chain=forward connection-state=invalid action=drop \
    comment="Descarta invalidas"
add chain=forward connection-state=new in-interface-list=WAN \
    action=drop comment="Bloqueia new da WAN"
add chain=forward src-address-list=bogons in-interface-list=WAN \
    action=drop comment="Drop bogons forward"
add chain=forward action=accept

# ===== NAT =====
/ip firewall nat
add chain=srcnat out-interface-list=WAN action=masquerade \
    comment="NAT Masquerade"
f) L3 Hardware Offloading
# ===== L3 Hardware Offloading no CCR2116 =====
# O switch chip Marvell Aldrin suporta roteamento L3 em hardware.
# Isso libera os 16 cores do CPU para firewall/queues/BGP.

# Verificar se o switch chip suporta L3HW
/interface ethernet switch print

# Habilitar L3 Hardware Offloading
/interface ethernet switch set 0 l3-hw-offloading=yes

# Configurar portas para offloading (exemplo com VLANs)
/interface ethernet switch port set ether2 l3-hw-offloading=yes
/interface ethernet switch port set ether3 l3-hw-offloading=yes
/interface ethernet switch port set ether4 l3-hw-offloading=yes
/interface ethernet switch port set ether5 l3-hw-offloading=yes
/interface ethernet switch port set sfp-sfpplus1 l3-hw-offloading=yes
/interface ethernet switch port set sfp-sfpplus2 l3-hw-offloading=yes
/interface ethernet switch port set sfp-sfpplus3 l3-hw-offloading=yes
/interface ethernet switch port set sfp-sfpplus4 l3-hw-offloading=yes

# Verificar status do offloading
/interface ethernet switch print detail
# Procure: l3-hw-offloading=yes e hw-offload-group

# Monitorar trafego offloaded vs CPU
/interface ethernet switch print stats
g) Queues PCQ
# ===== PCQ - Per Connection Queue =====
# Distribuicao justa de banda por IP de cliente

# Queue Types PCQ
/queue type add name=pcq-download kind=pcq \
    pcq-rate=0 pcq-classifier=dst-address pcq-burst-rate=0 \
    pcq-burst-threshold=0 pcq-burst-time=10s pcq-limit=50KiB

/queue type add name=pcq-upload kind=pcq \
    pcq-rate=0 pcq-classifier=src-address pcq-burst-rate=0 \
    pcq-burst-threshold=0 pcq-burst-time=10s pcq-limit=50KiB

# Mangle para marcacao de trafego
/ip firewall mangle
add chain=forward in-interface-list=WAN action=mark-connection \
    new-connection-mark=conn-clientes passthrough=yes \
    comment="Marca conexoes de download"
add chain=forward connection-mark=conn-clientes \
    action=mark-packet new-packet-mark=download passthrough=no \
    comment="Marca pacotes download"
add chain=forward out-interface-list=WAN action=mark-connection \
    new-connection-mark=conn-upload passthrough=yes \
    comment="Marca conexoes de upload"
add chain=forward connection-mark=conn-upload \
    action=mark-packet new-packet-mark=upload passthrough=no \
    comment="Marca pacotes upload"

# Queue Tree
/queue tree add name=download-total parent=global \
    packet-mark=download queue=pcq-download \
    max-limit=0 comment="Download PCQ"
/queue tree add name=upload-total parent=global \
    packet-mark=upload queue=pcq-upload \
    max-limit=0 comment="Upload PCQ"
h) Seguranca Avancada
# ===== Servicos =====
/ip service
set telnet disabled=yes
set ftp disabled=yes
set www disabled=no port=8080
set ssh port=2222
set api disabled=yes
set api-ssl disabled=yes
set winbox disabled=no

# ===== Hardening =====
/tool mac-server set allowed-interface-list=MANAGEMENT
/tool mac-server mac-winbox set allowed-interface-list=MANAGEMENT
/tool bandwidth-server set enabled=no
/ip proxy set enabled=no
/ip socks set enabled=no
/ip upnp set enabled=no
/ip cloud set ddns-enabled=no update-time=no

# ===== SSH Brute Force Protection =====
/ip firewall filter
add chain=input protocol=tcp dst-port=2222 \
    src-address-list=ssh-blocked action=drop \
    comment="SSH - Drop bloqueados" place-before=0
add chain=input protocol=tcp dst-port=2222 \
    connection-state=new action=add-src-to-address-list \
    address-list=ssh-stage1 address-list-timeout=1m
add chain=input protocol=tcp dst-port=2222 \
    connection-state=new src-address-list=ssh-stage1 \
    action=add-src-to-address-list \
    address-list=ssh-stage2 address-list-timeout=1m
add chain=input protocol=tcp dst-port=2222 \
    connection-state=new src-address-list=ssh-stage2 \
    action=add-src-to-address-list \
    address-list=ssh-blocked address-list-timeout=1d

# ===== Port Scan Detection =====
/ip firewall filter
add chain=input protocol=tcp tcp-flags=fin,!syn,!rst,!psh,!ack,!urg \
    action=add-src-to-address-list address-list=port-scanners \
    address-list-timeout=2w comment="FIN scan"
add chain=input protocol=tcp tcp-flags=syn,fin \
    action=add-src-to-address-list address-list=port-scanners \
    address-list-timeout=2w comment="SYN+FIN scan"
add chain=input src-address-list=port-scanners action=drop \
    comment="Drop port scanners"

# ===== Backup Automatico =====
/system scheduler add name=backup-diario interval=1d \
    on-event="/system backup save name=backup-auto dont-encrypt=yes" \
    start-time=03:00:00

# ===== Canal de atualizacao =====
/system package update set channel=long-term
i) BGP (Full Table, Peering, Route Filters)
# ===== BGP - Configuracao para ISP =====
# O CCR2116 processa BGP 6x mais rapido que antecessores.
# 16 GB de RAM comportam full table com folga.

# Routing table e ASN
/routing bgp template add name=default as=<SEU-ASN> \
    router-id=<SEU-ROUTER-ID>

# Peer com upstream (full table)
/routing bgp connection add name=upstream-1 \
    remote.address=<IP-UPSTREAM-1> remote.as=<ASN-UPSTREAM-1> \
    local.role=customer \
    templates=default \
    address-families=ip \
    multihop=no \
    hold-time=90 keepalive-time=30 \
    output.default-originate=never \
    comment="Upstream 1 - Full Table"

# Peer com upstream 2 (backup)
/routing bgp connection add name=upstream-2 \
    remote.address=<IP-UPSTREAM-2> remote.as=<ASN-UPSTREAM-2> \
    local.role=customer \
    templates=default \
    address-families=ip \
    multihop=no \
    hold-time=90 keepalive-time=30 \
    output.default-originate=never \
    comment="Upstream 2 - Backup"

# Peering (IX / PTT)
/routing bgp connection add name=peering-ix \
    remote.address=<IP-PEER-IX> remote.as=<ASN-PEER-IX> \
    local.role=peer \
    templates=default \
    address-families=ip \
    multihop=no \
    comment="IX Peering"

# ===== Route Filters =====
# Rejeitar prefixos invalidos e muito especificos
/routing filter rule add chain=bgp-in-filter \
    rule="if (dst-len > 24) { reject }" \
    comment="Rejeitar prefixos maiores que /24"

/routing filter rule add chain=bgp-in-filter \
    rule="if (dst == 0.0.0.0/0) { reject }" \
    comment="Rejeitar default route via BGP"

/routing filter rule add chain=bgp-in-filter \
    rule="if (dst in 10.0.0.0/8) { reject }" \
    comment="Rejeitar RFC1918"

/routing filter rule add chain=bgp-in-filter \
    rule="if (dst in 172.16.0.0/12) { reject }" \
    comment="Rejeitar RFC1918"

/routing filter rule add chain=bgp-in-filter \
    rule="if (dst in 192.168.0.0/16) { reject }" \
    comment="Rejeitar RFC1918"

/routing filter rule add chain=bgp-in-filter \
    rule="accept" \
    comment="Aceitar restante"

# Anunciar apenas seus proprios prefixos
/routing filter rule add chain=bgp-out-filter \
    rule="if (dst in <SEU-BLOCO/CIDR>) { accept }" \
    comment="Anunciar bloco proprio"

/routing filter rule add chain=bgp-out-filter \
    rule="reject" \
    comment="Rejeitar todo resto"
j) OSPF (Area Backbone entre POPs)
# ===== OSPF - Interligacao entre POPs =====
# Area 0 (backbone) para interconexao de CCR2116 entre POPs

# Instancia OSPF
/routing ospf instance add name=ospf-backbone \
    router-id=<ROUTER-ID> version=2

# Area backbone
/routing ospf area add name=backbone \
    instance=ospf-backbone area-id=0.0.0.0

# Interfaces OSPF (SFP+ para interligacao)
/routing ospf interface-template add area=backbone \
    interfaces=sfp-sfpplus3 networks=0.0.0.0/0 \
    type=ptp cost=10 \
    hello-interval=10 dead-interval=40 \
    comment="Link POP-1 via SFP+ 3"

/routing ospf interface-template add area=backbone \
    interfaces=sfp-sfpplus4 networks=0.0.0.0/0 \
    type=ptp cost=10 \
    hello-interval=10 dead-interval=40 \
    comment="Link POP-2 via SFP+ 4"

# Redistribuir rotas conectadas
/routing ospf instance set ospf-backbone \
    redistribute=connected

# Redistribuir OSPF no BGP (se necessario)
# /routing filter rule add chain=bgp-out-filter
#     rule="if (protocol ospf) { accept }"
Mevlox Distribuidora

Compre direto da distribuidora

Todos os nossos produtos sao homologados pela Anatel, com nota fiscal, garantia e envio para todo o Brasil.

Envio para todo Brasil
Nota fiscal garantida
Homologado Anatel
Suporte tecnico MikroTik
Precos para revendas
Falar pelo WhatsApp Enviar E-mail Visitar o Site

(86) 3142-6905 · vendas@mevlox.com.br · distribuidora.mevlox.com.br