I'm setting up a remote metro ethernet site (fiber in a closet) that
will have 2 x 100mb BGP transit feeds and a smattering of IGP feeds.
The traffic will be service provider transit without inspection, NAT
or other services.
Since everything is cost sensitive these days I initially planned on
implementing an ebayish 7206vxr-npe-g1. Although I was quite happily
slinging the 7206 around 10 years ago I realized tonight that it has
been 10 years and the 7206 platform is well aged. M7i (M7i 2AC 2FE
w/ RE400,PE-1GE-SFP) are quite common on the secondary market now and
likely more than enough to get started. Although trunking multiple
metro FE feeds to a single GE port will be frowned upon I may consider
this as an option.
I suppose my questions are whether a base M7i config out of the box
will support this application or if there are better options out
there. Thank you in advance.
_______________________________________________
juniper-nsp mailing list junip...@puck.nether.net
https://puck.nether.net/mailman/listinfo/juniper-nsp
The M7 is an awesome router for small to medium sites. It does have an
on-board GigE port, so if you can fit everything in that or a
downstream switch it could work.
However, it's really starting to show its age and there's not much
development happening on the M-series routers anymore (at least it
seems that way to me -- I'm sure they're still supported).
They're also pretty rock solid with JunOS 9. JunOS code quality and
feature-completeness has started to really slip since 10.0.
I'm not sure I totally understand from your description what you're
trying to build, but it sounds like you're looking for a router that
will support up to 200 Mbit/s of routed traffic that can speak BGP and
whatever IGP you're running.
If your environment is all copper ethernet (seems pretty common these
days), I might suggest checking out some of the nicer EX switches.
While really targeted at the "top of rack" market segment, they can
route up to 10GigE (with the right modules and platform), and speak a
variety of protocols (though some require extra software licensing).
With a little negotiating (remember, "list price" is very inflated),
you should be able to get a lot more bang for your buck over an older
M-series in an all-Ethernet environment.
My two cents.
Cheers,
jof
The application is a service provider edge, all ethernet, with routed
traffic to two carriers. Internal traffic is a mix of IGP and OSPF.
I'll have to take a look at the EX series. All of the literature on
the juniper site suggests the EX is targeted more toward lan
aggregation while the SRX handles the edge.
Thank you!
ex doesn't have enough fib for a ful table so If you need to take two
feeds and install all those routes, it's the wrong platform. m7i is just
ducky at the speed you're talking but the re-400 is a bit underpowered
and ramed for the modern era. re-850 with 1.5GB however is tollerable.
If your network is all ethernet and you don't plan on doing any TDM/SONET any time soon, I would look at the new MX80 bundles. With the right discount from your sales team, you can get an MX80 with 20 1G SFP-based ports for less than $20K. The MX80 has full internet route capabilities, 4 built-in 10G ports (although on the MX80-5G, they are "restricted", meaning you can't use them ;-)), and a "restricted" extra MIC slot. All these "restricted" options are enabled by a simple license purchase. The jury is still out on whether said restrictions are actually enforced, though - anyone have any experience with this?
The main problem with the M7i you listed is that the PE-1GE-SFP does not have per-VLAN queuing, which is becoming increasingly important in today's metro ethernet networks. The MX80 SFP ports also support 100M SFPs. You'd be much better off getting the MX80 than an M7i, if only for future-proofing your network. Yes, the M7i may be cheap on the secondary market, but if you plan on having this in production and getting software updates, you'll have to have it recertified by Juniper, which is something that can become quite costly.
-evt
Actually not all M7i's have the on-board GE, it depends on the BASE,
the base will either be M7iBASE-AC-2FETX which includes 2x 100mbit
copper Fast Ethernet ports on the inboard FPC, or M7iBASE-AC-1GE for a
single SFP gig-e port on board. These ports are seperate from the
100mbit management only port on the RE itself, you can NOT route
packets through the management port, it is only there to talk to the
RE, the RE can talk over it to export flows/etc, OR the RE can use any
of the PICs as normal. Those are AC power supply versions, there are
DC versions of same (that said I am pretty sure you can trade AC for
DC power supplies IIRC).
The M7i is a very solid platform itself, even though development is
slowing down, I kinda think the main reason for that is the platform
has pretty much reached all it can do. It can not support 10GE, the
forwarding plane/FPC complex just doesn't have the bandwidth. Even
the smallest CFEB shipped for the M7i has enough memory for full BGP
feeds. If you plan on feeding it a LOT fo full views you might
consider an E series CFEB
M7i PIC ports are wire speed (well, almost all Juniper M series ports
are, with a few exceptions of oversubscription in some configurations)
and will very handily push 200mbit of small packets even.
M7i and M10i are essentially the same router, the M10i has redundant
everything and four more PIC slots (on an extra FPC), the M7i only has
an option for a redundant CFEB.
Basically the ONLY time an M7i or M10i might not be able to do wire
speed is when you add services from the ASPIC or ASM (M7i only). And
if your'e not doing stateful firewalls or NAT (or a handful of other
time consuming not-exactly-router things) you'll never be able to hit
the limits on an M7i. The M10i if fully packed with Gig-E or other
highest speed ports can be marginally oversubscribed.
What was said later about EX series is true, if you don't need to
support anything but ethernet, and aren't doing advanced services,
it'd be a good fit for you, though they're still teething a little bit
(see other threads on this list).
It supports 20 x SFP Interfaces, came with ADC-R License , TRIO3D chipset
and 2GB DRAM (4m rib routes).
It came with 4 x XFP slots (blocked by software license)
On Thu, Mar 24, 2011 at 13:33, Doug Hanks <dha...@juniper.net> wrote:
> I would suggest the MX80.
>
> Doug
>
> -----Original Message-----
> From: juniper-n...@puck.nether.net [mailto:
> juniper-n...@puck.nether.net] On Behalf Of cjwstudios
> Sent: Wednesday, March 23, 2011 11:50 PM
> To: junip...@puck.nether.net
> Subject: [j-nsp] M7i
>
Doug
-----Original Message-----
From: juniper-n...@puck.nether.net [mailto:juniper-n...@puck.nether.net] On Behalf Of cjwstudios
Sent: Wednesday, March 23, 2011 11:50 PM
To: junip...@puck.nether.net
Subject: [j-nsp] M7i
MX80 Promotional 5G Bundle for channels, Includes MX80 Modular AC, spare
AC Power supply, 20x1G MIC including L3-ADV license, Queuing, Inline
Jflow, Junos WW. (4x10G fixed ports and 1x front empty MIC slot restricted)
This is a very good point, and one that I kinda didn't think about. It
would probably be fine to take a decently-sized IGP table, but not an
external one. Though it could be used to terminate an MPLS path to pin
the BGP sessions and traffic elsewhere.
There's kinda a hole in Juniper's product line between something small
like a J-series or SRX and an M or MX-series box.
I suppose the MX80 fills that hole somewhat, but certainly not
cost-wise. If you can work some aggressive pricing (which at the end
of a quarter or year can be easier), it can be a pretty good deal for
an amazing box.
If you can afford it, use an MX80 for an all-Ethernet environment.
I've got several going, and they're just great.
-Gabe
> M7i and M10i are essentially the same router, the M10i
> has redundant everything and four more PIC slots (on an
> extra FPC), the M7i only has an option for a redundant
> CFEB.
Ummh, not really.
The M7i eats only one CFEB at a time.
Mark.
> Hello Juniper folks :)
Sorry for the rather late chime-in on this one:
> Since everything is cost sensitive these days I initially
> planned on implementing an ebayish 7206vxr-npe-g1.
> Although I was quite happily slinging the 7206 around 10
> years ago I realized tonight that it has been 10 years
> and the 7206 platform is well aged.
Well, it depends on how you look at this.
The 7206-VXR chassis is really quite old, yes. However, the
NPE's are what keep the box going. While the NPE-G1 is quite
good, in many cases, it will top-out somewhere between
300Mbps - 500Mbps, depending on what you're doing.
The NPE-G2 should be able to touch 900Mbps with ease (I've
done 950Mbps on it - MPLS, RSVP, LDP, IS-IS, BGP, basic QoS,
IPv6).
Both the NPE-G1 and NPE-G2 are fairly modern if you consider
the fact that they are CPU-based forwarding boards. Also,
Cisco's IOS 12.2(33)SR* code base (now in its SRE iteration)
is very advanced. It's what keeps the box modern.
Even with all our M, T, MX, CRS, ASR1000 and ASR9000
platforms in the stable, the NPE-G1's/G2's are still very
handy, more so when nearly all connections are Ethernet in
nature.
The biggest problem with the 7206-VXR chassis is the
bandwidth points limitation, where certain PA's (port
adapters) will consume a certain amount of backplane
bandwidth, determining the limitation on how many PA's you
can scale to. If you looking at things like STM-1/OC-3,
Fast-E and Gig-E PA's, this can get hectic, but if you're
working on E1 PA's and things of the sort, there is nothing
to worry about. Slower interfaces don't generally tax as
much, if at all.
If you're only concerned about Ethernet, then the NPE-G1/G2
won't penalize you if you only use the ports on the NPE
itself. Of course, you quickly run into a density issue if
you need more than 3 ports. The 7201 will solve that as it
has 4 ports, but this may yet be still not enough.
For us, one of the biggest reasons we still maintain dozens
of 7206-VXR's with NPE-G1's/G2's is for situations where we
need to handle less than 1Gbps, but need a whole lot of
features which can come with restrictions in hardware-based
platforms, e.g., QoS, e.t.c. There are really are tons of
features in this platform's code that would put any hardware
box to shame - provided you can keep utilization below 1Gbps
for the NPE-G2. In low bandwidth peering sites, we'll
happily place an NPE-G1/G2 on the boat to go solve that
problem :-).
Will Cisco release an NPE-G3 and keep the platform relevant.
I don't know. I doubt, now that the ASR1000 is expanding in
cousins. But as long as the SR* code continues to be
developed for the 7200, the box will remain alive for the
simple reason that most folks that need to handle more than
1Gbps will not consider a 7200 anyway - but below that, it
is still a very viable option.
> I suppose my questions are whether a base M7i config out
> of the box will support this application or if there are
> better options out there. Thank you in advance.
The problem with the M7i is that it's now old. For the
amount of money the box costs, one doesn't feel they should
be restricted in the way the box does. If it were cheaper
(along with its components), I wouldn't complain as much.
Heck, I wouldn't complain at all.
That said, if you need hardware-based 6-in-4 tunneling
without worrying about buying an MS-PIC or Tunnel PIC, the
M7i is a great box for that. Then again, an NPE-G1/G2 will
do that for you quite easily, but in the CPU path (not quite
a bad thing unless you really need to handle more capacity).
Moving forward, though, if you need to handle all Ethernet,
I'd say consider the ASR1001/2/4/6 on the Cisco side, or the
MX80 on the Juniper side.
If you need to handle some Ethernet and some TDM/SDH/SONET,
I'd say consider the ASR1001/2/4/6 on the Cisco side, or the
M120 on the Juniper side.
Sadly, there is nothing in the Juniper portfolio that can
compete with the Cisco's ASR1000 line as of today. Waiting
for that book to be written :-). Meanwhile, we continue to
buy more ASR1000's for that role.
Cheers,
Mark.
Woops, you're most definitely right, I am not at all sure what i was
thinking about when I wrote that.
>
> Mark.