MKO ComNet Working Group Meeting Minutes


CFHT

April 12, 1995

Present:

Absent:

Minutes

  1. Network Status and Performance
  2. T-1 Characteristics

    Keck remote observers experienced during test runs that the T-1 link was not responding during certain applications. The problem was traced to the T-1 link itself which does not support B8ZS (Bipolar with eighth zero supression). As a result, the T1 link will experience problems when a string of 15 or more consecutive zeros is transmitted. This is a well known characteristic of T-1 lines also known as the one-density requirement.

    After studying various ways of fixing this problem, we have switched the CSU/DSU to HDLC (High-level Data Link Control) with simple D4 framing and at a line speed of 1.53 Mbits. HDLC will invert the data being sent, thereby reducing the possibility that a string of 15 or more consecutive zeros will be sent. It is possible, but less likely that data would include 15 consecutive ones. We will monitor the performance to determine if this arrangement is satisfactory until the fiber from HP to Humuula is installed, at which time B8ZS will certainly be available.

    Other options include relocating the T-1 line to the summit microwave which does support B8ZS, or asking Hawaiian Tel to upgrade its radios. The last option is very unlikely, as the costs to upgrade the microwaves at HP and Mauna Loa are about $100,000.

    FDDI Ring

    The FDDI ring had two outages to report. One involved a burned out power supply at JCMT, the second an overheated router at CFHT. Unfortunately they occurred at approximately the same time resulting in the isolation of CFHT and Keck from the Internet for a short time. The CFHT and the JCMT routers have both been restored. However the FDDI ring is wrapping at KECK and JCMT, a condition which will be remedied within a few days.

  3. MKO Communications Infrastructure Improvement Plans
  4. Steve Hardash presented an updated construction plan for Gemini for the remainder of this year. San Juan Constructions has completed 80% of the road relocation work. It is anticipated that the utilities relocation will start the week of April 17. Their first order of business is to remove the old 24" infrastructure. The new conduits will then be installed. HELCO, HawTel, and the fiber optics cables will need to relocated before a majority of the Gemini foundation work can begin. Some minor work is anticipated to start in May. The fiber relocation is projected for the late June to early July timeframe. The contractor chosen is Kailua Kona Resort Corporation with T&T Electric the subcontractor and Rich Bacon the consultant.

    The detailed infrastructure relocation plan has been discussed previously. The splicing is planned to be performed at the new handhole close to HH16. Each fiber optic cable, starting with the unused Subaru fiber sheath, will be cut at HH15, pulled back individually to HH16 and pulled into the new handhole where the splicing will take place. This procedure is projected to take less than 6 hours. As a precautionary measure an access hole about 50 ft south of HH16 will be dug up, and the existing conduit cut in case it is not possible to pull an individual fiber all the way back from HH15. In that case the fiber will be cut at the access hole and pulled back to HH16.

    We are also exploring the possibility of providing a 28.8 kbaud PPP connection from the telescope facilities/base to Hale Pohaku for access to Internet as a backup measure. Details will be forthcoming.

    Phase II which was rewritten since the last meeting to avoid laying a temporary surface cable was reviewed. It was decided that the following information should be provided by each facility in time for the next meeting:

    1. Possibility/desire of replacing fiber from each facility to the handhole,
    2. Congestion of the conduits leading into each facility,
    3. Distance between the handhole and termination point within each facility.

    Each facility will investigate the latter two points and forward the information to IfA, in order for the IfA to come up with a cost estimate of new fiber pull compared to re-using the old fibers.

  5. Interop Summary
  6. The focus of Interop has shifted from ATM everywhere to Ethernet switching for a LAN. A summary of current ethernet switch vendors is available. One concern which was brought up was the corporate health of Proteon. Even though Proteon has limited their development of high end routers, their financial position is strong. Since we are not in the market for router upgrades at the moment, it is not really an issue. The technology may have changed significantly by the time we are ready for an upgrade at which time we will reevaluate our hardware supplier.

    Currently and forthcoming interesting conferences include Internet World, International WWW Conference, INET 95.

  7. Other Business
  8. Resplicing of Faulty Fibers

    After extensive discussion, it was decided that IfA will look into contracting out to Telecable Systems in Kona to resplice the fibers which were determined faulty recently. The facilities will be recharged for their share of resplicing, which include only those within their telescope. Resplicing of fibers at the Summit Communications Room will be paid for by IfA.

    Facilities will be asked to approve the cost or issue a purchase order before the work for their location is contracted.

    Interdome Fiber Link

    Andrew Pickles indicated the desire to have fiber communications to the remaining 24" telescope and to CFHT. The use of the spare single mode fibers between 88 inch and CFHT is a possibility. It would require the use of single-mode to multi-mode converters. Another possibility is to install subducts in the new communications duct. Steve Hardash will explore the cost involved.

  9. Next Meeting
  10. Friday July 14, 1995, 10:00am at IfA (subject to change).

MKOCN Homepage Back to Minutes

Last updated December 7, 2001 by Miranda Hawarden-Ogata