Legacy ISDN over IP

Discussion in 'Royal Signals' started by PoisonDwarf, Jan 3, 2010.

Welcome to the Army Rumour Service, ARRSE

The UK's largest and busiest UNofficial military website.

The heart of the site is the forum area, including:

  1. Man, this forum is getting dull. We need more technical stuff or the other Corps will think all we do is bitch-slap each other about trades.

    I was thinking to myself (sad as it seems) that it's disppointing how little serious progress we have made in migrating to all-IP networks in our mob. I watched three RAF bods recently spending about 15 mins jacking up a VTC and thought surely there's got to be a better way of engineering stuff. They were doing what I saw lads in 30 Sigs doing nearly 10 years ago and with a virtually identical system!

    There are a number of dodgy civvy companies (the usual cartels) who make an awful lot of money ripping off MoD plc with their over-complicated but architecturally archaic solutions that require vast amounts of contractor support. Although we are often hamstrung by the system, I think that we often don't help ourselves and it's just far too easy to accept the so-called professional opinion of these big companies. It's about time that the military regained the initiative! (yeah I know that's not going to happen - there's too much money in contractors prologing the agony).

    Anyway, enough rabble-rousing. I caught sight of a piece of kit called a Packetband and wondered if anyone has ever encountered it? It looks like a very nice little bit of kit - defo ripe for an in-house eqpt trial and (from the link) it looks like it is probably in-service somewhere, but I don't know where. I don't want to reinvent the wheel - if anyone can tip me the nod I'd be grateful. I'll share my inevitable OBE with you. :lol:
  2. PM me.
  3. Roger that. Check PMs.
  4. Why setup a legacy ISDN link over what is likely to be an already secure IP connection? There are means for secure voice and data/video over IP already without the cost overhead of the unit. I hope the full Unified Comms capability of the platform being rolled out on F (and FD???) will be used.

    Would be good for lab use though.

    Contractors are needed as unfortunately skills retention/fade/management in the MOD is laughable. There are countless networks setup by a very IT literate bunch of Sigs (or even other Corps!!) with non standard skill sets (so MCSE, CCNA equivalent with nouse behind it) who were then posted/deployed/demobbed and replacements brought in who barely knew their USB port from their Ethernet. My small team has managed/developed/supported a small MOD system for the last 8 years - none of the original users and managers are left and we are constantly training the new folks and guiding new managers in order to make sure they get the best use from the system (and ultimately make us richer). However none of the MOD bods have taken a direct interest in managing/supporting the capability themselves, due to a lack of time, manpower or skills. It is just more efficient to use us.
  5. Doesn't net.com have a solution, the problems were usually the stability of the link (country to country) not the vtc gear or equipment.
  6. Dont let the IDA Nazis catch you :)
  7. I'm reluctant to be specific on kit, even if it is commercial stuff referred to on the link on my original post, but there is certain stuff that we always roll out in virtually every ex / op that remains ISDN-based, including voice and video. The fact is that we (MoD plc) are still miles behind the curve and it'll be some time before we end up with unified comms as the standard.

    I think your points are all valid and it highlights the limitations of designing systems around vendor specs per se. What we should be concentrating on is open standards, so it doesn't matter if the kit is from Cisco, Nortel, Juniper or the Disney Corporation - they all use the same language / standard / process / protocol. We should exploit the technologies, rather than being their victim. Inevitably, far too many guys involved on important developments in technology are all senior officers who have never really been involved in CIS / ICS and have just completed some 2 year Shrivenham MSc before going to the IPT or DE&S. Then what happens is that they don't understand the challenge, don't have a handle on professional / industry developments and certainly can't identify the solution, so they allow contractors to come in and define their own solution. And so we perpetuate the old "well, that's what you asked for", which is a waste of taxpayer money and an insult to professional communicators everywhere.

    The problem with relying on contractors (and I know I'll probably be one before too long) is the same as with drug companies. There's no money in the cure - the smart money is on prolonging the treatment.

    Great news for ex R SIGNALS everywhere though! :lol:
  8. PD, you deserve recognition for the number of questions and comments raised in a single post, but I'll try and answer them seperately. In no particular order:

    1. The Packetband has been trialled and used in a number of applications and locations. As Chief Muppet points out, pushing ISDN over an IP network is somewhat self-defeating and so its use is limited but a typical application would be to extend a single service over someone elses (eg US) infrastructure. If you want further information about the type of application where it is a suitable solution then I believe that they have been used in a number of FofS (non-IS) projects over the past few years, details are available from the library at Blandford.

    2. One of the major advantages of using an ISDN network is to retain standardisation with the UK/Germany end. Trying to move the deployed network to entirely IP whilst DFTS is running on ISDN would introduce another level of complexity into networks which are already overly complicated.

    3. IP isn't the panacea which some seem to believe that it is. There are a number of limitations to IP networks, particularly when high latency links (ie sat links) are being used. Given our reliance on satellite it is questionable whether an all IP network would be of benefit. Work has started however to move parts of the Herrick network over to IP.

    4. I was a little confused by your comments about the usual cartels of dodgy civvie companies. I fully agree with your sentiments but I didn't see just how they applied until I remembered that you are talking about Telic and not Herrick. The network in Herrick, which is now our main effort, is fully owned, maintained and operated by us and not Paradigm. I'll admit that the equipment and technology is actually older than Paradigm's but it works and is, given the environmental limitations, both robust and reliable. My own opinion is that the problems with the Herrick network are not a result of the equipment or technology but rather the clusterfuck that DES have made of the management. The main, unspoken, advantage of using old and outdated equipment is that to replace the entire network would require a PFI and handing over the control to the aforementioned usual cartels.

    I'm going to finish for now as the G&T's are starting to take effect. The above comments are deliberately vague due to the limitations of replying on a public forum.

    I realise that parts of my comments are contradictory, this is partly due to avoiding specifics but mainly due to the joys of trying to maintain a network in a military/DES environment.
  9. for technical aspects:

    Google IPMUX ISDN

    There is a wide range of the equipment available to send TDM over IP. using this technique I have had a remote GSM basestation in south america and the BSC and core network were in the UK. It was used to demonstrate handovers between WIFI and GSM (google unlicenced mobile access UMA or generic access network GAN).

    It is easy to setup and run.

  10. Deejay, good call about the FofS projects and some good stuff about IP etc. I shall get in contact with WO1 bloke at RSS about Packetband and similar kit.

    Biggest problem with a unified packet-switched network is applying the right QoS to the right stuff and I reckon that's the reason that we're still sat with so many stovepipes. It's in the "all too difficult" pile and even in the Corps we don't always see people are actively seeking out newer, better ways of working. They see obstacles rather than opportunities, but DTA is coming, so it can't be ignored forever. I like the internet IP/MPLS stuff going on.

    Ref Herric, Telic etc, I don't think I referred to any particular theatre in my post. Although my interest was stimulated by something specific, I am looking at the general technique and how we can get quick wins in stovepipe systems.

    The cartels I refer to are pretty much all of them. We might own the Herrick network now, but it's only a matter of time before the firm sells it off as an outsourced / managed service over Skynet surely? The IT stuff is the same - it might be military now but it will all migrate to Atlas as soon as the risk has reduced and they can see the profit. More money for EDS and friends.

    Cheers for a good post as always.
  11. 0900 from Kings X, she could chat with her grandson on the way up (plus it's far cheaper, far less organisation and far more likely to arrive on time due to the lack of bullshit).
  12. Roadster, check you out. You must be knackered after that. :D
  13. ISDN over IP? Which is TDM over a packet network.
    Can be done using an PWE3* service over IP/MPLS - but will be expensive and complex.

    Why over IP? What is wrong with TDM? Are you just after a cheap voice and data service?

    *Psuedowire End to End Emulated Service
  14. I concur with The Eagles.

    "You can check out any time you like, but you can never leave"
  15. PD, check your PMs.