From its websites, we could know EOS is built on top of Linux and written using C++ and Python. L2/L3 network protocols, CLI, SNMP, and ASIC drivers are running in user space. ProcMgr is the process to do stateful restart and recovery based on System Database (SysDB), which works as IPC mechanism between different user agents. The innovation of EOS is its vEOS, VM Tracer and MLAG.
In big picture, Arista Networks shares similar view as Cisco's vision on data center network: Access, Aggregation and Core. This may came from the fact that Arista's CEO and founders ever worked in Cisco Systems. The performance looks promising from disclosed data sheet. The per-port pricing is attractive too. However, compared to Cisco's portfolio, the choice Arista Networks provided is limited. The feature set is not that rich. What's more, the solution between leaf switch and server or VM is software-based. Not sure whether it would affect the performance when system scaled bigger.
The following lists switches they provided today.
- Arista 7048 (Leaf)
- 1 RU
- 48 1000Base-T ports
- 4 1/10GbE SFP+ uplink ports
- 2 redundant hot-swappable power supply
- 4 redundant hot-swappable fans
- 768 MB deep buffer
- Citrix NetScaler VPX integration
- L4-L7 Load Balancing & Security
- Arista EOS
- Arista 7010 Series(Spine)
- 1 RU
- 24 or 48 10Gbs SFP/SFP+ ports
- 2 redundant hot-swappable power supply
- 5 redundant hot-swappable fans
- 2 10/100/1000Mb Mgmt ports
- 1 RS-232 serial ports
- Arista EOS
- Arista 7508(Core)
- 11 RU Chassis with mid-plane design
- 48 * 8 = 38410Gbs SFP/SFP+ ports
- 4 redundant hot-swappable power supply
- 6 fabric modules
- 2 supervisor modules
- 6 fan modules
- Arista EOS
No comments:
Post a Comment