可成长软件理论方法和实现技术:从范型到跨越  

Theories and techniques for growing software: paradigm and beyond

作　　者：许畅[1,2] 秦逸[1,2] 余萍[1,2] 曹春[1,2] 吕建[1,2] Chang XU;Yi QIN;Ping YU;Chun CAO;Jian LÜ(State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210023,China;Department of Computer Science and Technology,Nanjing University,Nanjing 210023,China)

机构地区：[1]南京大学计算机软件新技术国家重点实验室,南京210023 [2]南京大学计算机科学与技术系,南京210023

出　　处：《中国科学：信息科学》2020年第11期1595-1611,共17页Scientia Sinica(Informationis)

基　　金：国家重点研发计划(批准号:2017YFB1001801);国家自然科学基金(批准号:61932021,61902173)资助项目。

摘　　要：在云计算和大数据的技术背景下,"人–机–物"三元融合的应用模式正不断加速社会的信息化进程,并对软件系统的自适应和持续演化能力提出了新的需求.本文探索了面临软硬件环境及外部资源不断变迁挑战下的可成长网构软件理论方法和实现技术,从软件可成长性问题的由来,至可成长性概念的内涵和可成长软件的范型机理,在开放环境感知与自适应、无缝演化和过程演进,以及演化质量评估方法和保障机制3个方面系统分析了可成长软件的技术挑战并介绍了当前的技术进展,以支撑软件系统在不断成长视角下的长期生存.With advances in cloud computing and big data technologies,"human-cyber-physical"applications are providing increasingly rich information and robust functionality.This imposes new technical changes on software systems,which are required to make self-adaptation and continuous evolution to meet our increasingly higher expectations.This article explores theories and techniques for growing software to meet the challenges caused by constantly changing environments and external resources.It studies the source of the software growability problem,seeks to define software growability,and develops a paradigm for growing software.It further analyzes the challenges of supporting environmental sensing and self-adaptation,realizing seamless evolution and process optimization,and developing quality evaluation and assurance mechanisms for growing software.It also reports recent technical advances in these areas from the perspective of long-living and continuously-growing software.

关 键 词：可成长软件 范型机理 自主适应 持续演化 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]