Hi Qiao and all, I reviewed the latest exascale network optimization draft. In addition to the review, the key design questions I want to raise discussions are (1) How to describe DAG of workflow converter in more detail？ Could we give some examples for different DAG? (2) We give some user API in section 7.2.1, Could we provide some JSON format for it? (3) In the query mode, we also need some JSON string to describe it.
If you have any comments or suggestion, please share with me. Thank you so much. In addition, please see my other comments in the following. *Page 4* Production ANd Distributed Analysis system (PanDA) in ATLAS and the Physics Experiment Data Export system (PhEDEX) in CMS have been *[HZ] ”AND” -> ”and”* *Page 5* [DRAFT-CC] and etc. to encode information about multiple types of resources in science networks, *[HZ] ”to” -> ”To”* *Page 6* Different sites in this network belong to different administration domain. *[HZ] ”domain” -> ”domains”* the orchestrating framework must be able to collect sufficient resource information about different clusters/sites in real-time as well as over the longer term, *[HZ]”must”->”MUST”* *Page 7* Hence the orchestrating framework must be modular so that it can support different dataset management systems and different orchestrating algorithms. *[HZ]”must”->”MUST”* The orchestrating framework must support the interaction between the multi-resource orchestration module *[HZ]”must”->”MUST”* abstraction, multi-cost, cost calendar, etc. have been proposed to provide fine-grained network information to applications. *[HZ]”etc.” -> ”etc.,”* *Page 10* a large dataset transfer or analytic application always involve many network elements in multiple clusters/sites and the absolute number of involved network elements keep increasing as the scale of clusters increase. *[HZ]”involve”->”involves” * *Page 12* STEP 6 The ALTO servers at each site accept the queries from the ALTO client, collects resource information from the residing site and sends back to the ALTO client. *[HZ]”collects”->”collect”, “sends”->”send”* *Page 13* and the relative priorities may be modified by the resource orchestrator depending on the role of users (regular users or administrators at different levels), *[HZ]”may”->”MAY”* *Page 14* The ALTO client should operate in different query modes depending on the implementation of ALTO servers. *[HZ]”should”->”SHOULD”* *Page 15* Each ALTO server must provide basic information services as specified in [RFC7285] such as network map, *[HZ]”must”->”MUST”* *Page 16* The multi-resource orchestrator should adjust the resource allocation decisions based on the progress of ongoing requests, the utilization and dynamics of cluster resources. *[HZ]”should”->”SHOULD”* If we have any mistake, please feel free to figure out. Thanks a lot! Best Regards, ************************************************************* Haizhou *************************************************************
_______________________________________________ alto mailing list [email protected] https://www.ietf.org/mailman/listinfo/alto
