WIS data-communications functions and services, and implementation Progress report on Expert Team on WIS-GTS Communication Techniques and Structure (ET-CTS) ICG-WIS-4 (Reading, 4-7 September 2007) Hiroyuki ICHIJO (Japan)

ICM-IMTN (cloud I) (October 2007) Progress Report Interim Report ET-CTS-1 (Tokyo, April 2006) ICG-WIS-3 (Beijing, 5-8 September 2006) Follow-up discussion Outcome Compiling further study items ICG-WIS-4 (Reading, 4-7 September 2007) Discussion and finding solutions ICM-IMTN (cloud II, Africa) (1Q 2008) ET-WISC (4Q 2007, 1Q 2008) Contribution ET-CTS-2 (2Q 2008) ICG-WIS-5 (3Q 2006) Final ReportRecommendation ICT-ISS (3Q 2008) CBS-XIV (4Q 2008) Position of this report

I. Approach models to design for WIS data communication structure Totality approach model Assembling approach model Further study items Design principles Solutions Design principles Solutions Design requirements Most feasible design

1. Totality approach model Study steps are: i)create a global any to any network (e.g. Global MPLS Cloud); ii)provide an IPVPN backup solution keeping any to any connectivity; iii)on top of the any to any network connection map the needed application connection. Site D acts as a gateway for Site F Site A acts as a gateway for Site G Consideration points are: i)how to connect with WIS participants other than NMHSs; ii)administrative aspects such as coordinated contract, management and governance; iii)gateway issues such as responsibility, the expected number and technical requirement.

2. Assembling approach model The team has been discussing design principles of individual components based on this model so far. The reasons are: i)differences between components in migration target period, feasibility and accuracy of implementation plans, the number of related centers, required specifications, design flexibility and coordination difficulties in administrative aspects; ii) importance and requirements of initial establishment of a WIS core network; iii) understandability and easiness of practical study and smooth evolution. Consideration points are: i)how to harmonize all components as a WIS data communication structure; ii)administrative aspects in each component; iii) how to shorten long term evolution period.

II. Further study Items with a high priority 1. Concepts for the WIS core network 1.1 Backup for a WIS core network GISC WIS core network Internet GISC WIS core network #1 WIS core network #2 GISC (1) Internet backup (2) Dual network configuration Consideration points: i)Backup control, automatic or manual i)Security requirements and sustainable level ii)Limitaion of backup traffic Consideration points: i)Risk avoidance by two suppliers ii)Cost tolerable condition iii)Usual traffic load balancing iv)Administrative and technical aspects in implementation

1.2 Backup connections with alternate GISCs Team’s goal is to recommend necessary backup cases with appropriate options by alternate GISCs. Consideration points are: i) Coordination issue between GISCs but consideration from the NC view ii) Backup traffic scenarios for time-critical operational data, near real- time products and ad hoc Example configuration and backup case NC GISC#1 GTS Alternate GISC#2 Internet VPN Backup case Internet GISC-NC connection components GISC NC Internet Portal Internet access to a portal for ad hoc request/reply Internet VPN link for near real-time products VPN GTS for time-critical operational data

2. Smooth evolution and migration scenario 2.1 Evolution process from the Improved MTN to the WIS core network Consideration points for contribution to ICM-IMTN meetings i) Reviewing outcome of the ET-CTS-1 Use of closed network cloud(s) by SLA based service Limitation of the number of GISCs for practical reasons Benefit of full-mesh topology in case of a small number of GISCs ii) Data synchronization on application connections Full-mesh or non-full-mesh with sophisticated Data GRID (e.g. SIMDAT) Unicast or multicast-oriented architecture Future possibility to use IPv6 multicast iii) Current IMTN configuration of two clouds GISC Unicast-oriented network GISC Responsibility Area GISC Multicast-oriented network GISC Responsibility Area Multicast group Duplicated transmission

Current IMTN configuration Exeter Sofia Melbourne Buenos Aires Tokyo Beijing Nairobi Washington Cloud Cloud I Jeddah Prague Toulouse Dakar Algiers Offenbach Brasilia Cairo New Delhi Moscow Cloud II

2.2 Adjustment of the current GTS configuration to WIS topology The team will study necessary adjustment considering the following points: i) Mapping and hierarchical design: RTH(MTN) – RTH(non-MTN) - NMC … GISC – DCPC – NC ii) Analysis of shortcomings of the current GTS 3.Standardization of protocols and procedures The team will contribute technical recommendations to regulatory documentation. Possible contribution items are: i) Comparison of VPN techniques ii)Differences between WMO standards and industrial standards iii)Standard VSAT Furthermore, an important aspect is the definition of the governance structures to control data flows and growth of traffic. In a new WIS core network, telecommunication becomes ubiquitous and general purpose. The telecommunication managers may not be able to see and control data flows and growth. Governance of data exchange needs to be redefined.

4. Administrative aspects In spite of their difficulty and sensitiveness, the following administrative items should be studied in cooperation with other teams and/or groups such as ICM-IMTN groups and the IGDDS implementation group. i) Hurdles in administrative aspect of the collaboration framework by multi-nations ii)Adaptability to contracts on SLA and agreement on MoU iii)Appropriate collaboration schemes for satellite-based communication systems Example of a collaboration framework to join a common managed network service Managed Network by a single provider GISC DCPC NC Collaboration Framework MoU Contract with SLA

5. Technical solutions of key issues 5.1 Complementary Internet VPN links for overflow traffic In spite of their difficulty and sensitiveness, the following administrative items should be studied in cooperation with other teams and/or groups such as ICM-IMTN groups and the IGDDS implementation group. 5.2 Effective load balancing on dual network configuration i) Hurdles in administrative aspect of the collaboration framework by multi-nations ii)Adaptability to contracts on SLA and agreement on MoU iii)Appropriate collaboration schemes for satellite-based communication systems From a network point of view, the team has not found a firm solution on load balancing another network. 5.3 Possible services for mobile terminal users GISC DCPC Closed network Internet

Thank you for your attention and valuable comments On behalf of ET-CTS, one of the most powerful teams