• Aucun résultat trouvé

ERP B3: Service Level Driven Management of On-Demand Business Support Systems

N/A
N/A
Protected

Academic year: 2022

Partager "ERP B3: Service Level Driven Management of On-Demand Business Support Systems"

Copied!
3
0
0

Texte intégral

(1)

ERP B3: Service Level Driven Management of On-Demand Business Support Systems.

?

Ulrich Winkler1, Daniel Playfair1, and Wolfgang Theilmann2

SAP Research, SAP AG,

1The Concourse, Queen’s Road, Belfast BT3 9DT, United Kingdom

2Vincent-Priessnitz-Str. 1, 76131 Karlsruhe, Germany {ulrich.winkler,daniel.playfair,wolfgang.theilmann}@sap.com

http://www.sap.com/research

Abstract. ERP B3 is a SLA@SOI framework based solution for Ser- vice Level Agreement driven, on demand and dynamic provisioned ERP systems. In this demonstration we want to show SLA management for hosted ERP systems from three perspectives; the customer, the sales, and IT administrator perspective. Aspects of the negotiation, planning and provisioning workflow, which involves all stakeholders, are outlined.

1 Introduction

Business support systems, such as Enterprise Resource Planning (ERP) sys- tems are well established in large organisations and hosted on customer premise.

However, the uptake for small and medium enterprises is still low due to high- complexity and high initial-costs of setup and maintenance. To offer ERP func- tionality as a Software-as-a-Service (SaaS) offering could be a solution. As ERP systems are business critical systems, customers should be enabled to specific Ser- vice Level Agreements (SLAs) which details penalties in case the EPR provider is not able to deliver ERP functionality as demanded.

The SLA@SOI project researches and develops a framework for SLA aware SaaS solutions [1, 2]. This framework provides (1) comprehensive support for holistic and transparent SLA management, SLA translation and SLA negotia- tion, (2) a means to predict service quality characteristics, (3) an automated service deployment apparatus and (4) mechanisms to monitor and to enforce service quality at runtime.

Lessons learned from utilising the SLA@SOI framework to provide on-demand business applications are discussed in [3]. The authors elaborate on details how the SLA@SOI framework is used to plan, translate and negotiate SLAs at dif- ferent layers and explain technical and scientific particulars. However, the best way to illustrate this lengthy end-to-end planning, translation and negotiation workflow is with a demonstration.

?The research leading to these results is partially supported by the European Com- munity’s Seventh Framework Programme (FP7/2001-2013) under grant agreement no.216556.

(2)

2 Ulrich Winkler et al

2 ERP B3’s SLA Negotiation and Translation Workflow

We anticipate three stakeholders in the negotiation and planning workflow; which are the customer, the sales officer and the IT administrator. For every stakeholder ERP B3 offers a tailored user front-end, called portal. The customer portal al- lows a customer to browse product offerings. The customer can initiate a quo- tation process and specify service level requirements. The sales portal provides functionality to manage customer requests, to plan business service level agree- ments and to perform price calculations. The IT administrator portal supports IT landscape planning and provide monitoring and adjustment functionality.

Fig. 1.The negotiation, planing and provisioning workflow.

(3)

ERP B3 3

The end-to-end negotiation, planning and provisioning workflow is depicted in Figure 1 and briefly discussed here:

The customer [steps 1-5.1] browses the product catalogue and select a product of interest. For every product the customer can choose from various pre-defined SLA templates, configures the selected template according to his business needs, and issues a request for quotation to the sales office.

The sales officer [steps 5.1.1-7] examines quotation request details and prop- agates the IT relevant parts to the IT administrator as an application service request.

The IT administrator [steps 7.1-9] receives the application service request and triggers the IT planning wizard [step 9 ff]. This wizard translates appli- cation service requests into a set of IT landscape plans, including middleware and infrastructure requirements. The IT administrator [steps 10-10.1] selects the most appropriate landscape plan. Once an IT landscape plan has been selected the sales officer can finalises the quote for the customer, i.e. she determines a final price.

The customer [step 13] accepts the quote. This triggers an automated pro- visioning process for the complete service hierarchy. The sales officer and IT administrator are notified accordingly.

3 Conclusions

Server virtualisation and cloud computing enable new kinds of service provi- sioning and management methodologies. The SLA@SOI framework supplements these methodologies with SLA management, which is required in a business context. ERP B3 makes use of this framework to provide service level aware on-demand business applications. In this paper we have shown the end-to-end negotiation, planning and provisioning workflow, which co-ordinates SLA man- agement activities among various stakeholders.

References

1. SLA@SOI project: IST- 216556; Empowering the Service Economy with SLA-aware Infrastructures,http://www.sla-at-soi.eu/

2. Theilmann, W., Happe, J., Kotsokalis, C., Edmonds, A., Kearney. K.: A Reference Architecture for Multi-Level SLA Management. Journal of Internet Engineering, 2010 (to appear)

3. Wolfgang Theilmann, Ulrich Winkler, Jens Happe, and Ildefons Magrans de Abril:

Managing on-demand business applications with hierarchical service level agree- ments. 3rd Future Internet Symposium (FIS 2010), Berlin, September 20-22, 2010

Références

Documents relatifs

• Organize LKMS parts (adding resources libraries and handlers to the LKMS model): knowledge handlers and documents (ontologies, metadata, etc.); primary resources (documents, tools,

infrastructure capable of providing full cloud capabili- ties. In an IaaS environment, a cloud provider acquires a hardware infrastructure such as main frames, servers, networks

Cloud Computing QoS is a very important aspect as a customer will only expe- rience the launching and the resuming of a service provisioning request regard- less how many

Everything starts with two measurement directives given by the WSNoperator Information System: the System Generated Message Counter (SGMC) which is the number of messages generated

As discussed before, adaptation can happen either on instance (level 1, i.e., adaptations which affect only a single instance) or structural level (level 2, i.e., adaptations

• Sub-process A: Initially, the monitoring framework extracts the guarantee terms from the SLAs related to SBS provisioning and translates the guarantee terms of these SLAs into

This involves: (i) understanding the dependencies between KPIs of the process and performance characteristics of orchestrated partner services and used IT infrastructure;

Different types of knowledge are encountered in IT domains: product knowledge, procedural knowledge, legal knowledge, behavioral knowledge, customer knowledge and topological