PROPOLIS：数据驱动过程的假设分析系统

需积分: 10 72 浏览量更新于2024-09-09 收藏 345KB PDF 举报

"PROPOLIS系统提供了对数据驱动流程的预配置分析，允许分析师在假设的修改场景下探索和测试在线购物等应用的效果。该系统基于数据来源的概念，通过紧凑的表达式捕获所有可能场景组合的分析结果，并支持快速交互式探索。" 在IT领域，数据驱动的流程应用广泛，例如在线购物、银行交易和库存管理等，这些系统依赖于底层数据库并由有限状态转换系统控制其行为。"Provisioned Analysis of Data-Centric Processes"这个概念针对的就是这类应用的分析需求。传统的分析方法通常关注于单一实例的流程控制和数据库状态，而PROPOLIS则超越了这种限制，它允许用户进行动态的、假设性的修改，以便理解不同改动如何影响系统的分析结果。 PROPOLIS系统的核心是“预配置表达式”，这一概念源自数据来源（Data Provenance）的理论。数据来源追踪数据从生成到使用的路径，而预配置表达式则是这一概念的扩展，它能够以一种压缩的形式捕获所有可能的假设场景组合对分析结果的影响。这意味着分析师可以快速地模拟和测试多种场景，而无需逐一执行每个场景来获取分析结果。在演示中，PROPOLIS将应用于在线购物应用程序，参与者可以扮演分析师的角色，实时探索如价格调整、库存变化或用户行为更改等假设情况对整个流程分析的影响。这不仅提高了分析的效率，也使得理解和优化这些复杂数据驱动流程变得更加直观和便捷。通过PROPOLIS，我们可以实现以下几点： 1. 动态场景探索：用户可以快速评估不同修改对流程性能、安全性和合规性的影响。 2. 高效分析：预配置表达式减少了计算复杂度，使交互式分析成为可能，即使面对大量可能的场景组合。 3. 决策支持：系统提供的即时反馈帮助决策者更好地理解潜在变更的后果，从而做出更明智的业务决策。 4. 风险评估：在实际实施变更之前，可以通过模拟来预测潜在风险和问题，降低操作风险。 5. 流程优化：通过对各种假设场景的分析，可以识别瓶颈和改进点，提升流程效率。 PROPOLIS为数据驱动的流程分析提供了一个强大而灵活的工具，它促进了对复杂系统行为的理解，增强了分析的深度和广度，有助于在业务环境中实现更好的决策制定和流程优化。

PROPOLIS: Provisioned Analysis of

Data-Centric Processes

∗

Daniel Deutch

Ben Gurion University

Yuval Moskovitch

Ben Gurion University

Val Tannen

University of Pennsylvania

ABSTRACT

We consider in this demonstration the (static) analysis of

data-centric process-based applications, namely applications

that depend on an underlying database and whose control

is guided by a ﬁnite state transition system. We observe

that analysts of such applications often want to do more

than analyze a speciﬁc instance of the application’s pro-

cess control and database. In particular they want to in-

teractively test and explore the eﬀect on analysis results

of diﬀerent hypothetical modiﬁcations applied to the appli-

cation’s transition system and to the underlying database.

To that end, we propose a demonstration of PROPOLIS, a

system for PROvisioned PrOcess anaLysIS, namely analy-

sis of data-centric processes under hypothetical modiﬁcation

scenarios. Our solution is based on the notion of a provi-

sioned expression (which in turn is based on the notion of

data provenance), namely an expression that captures, in a

compact way, the analysis result with respect to all possible

combinations of scenarios, and allows for their exploration

at interactive speed. We will demonstrate PROPOLIS in the

context of an online shopping application, letting partici-

pants play the role of analysts.

1. INTRODUCTION

Many real-life applications rely, in intricate ways, on an

underlying database in their operation. For example, E-

commerce applications such as that of Ebay [9] rely on a

database for management of products, orders etc., which

aﬀects the possible execution of the application and its in-

teraction with potential users. Due to the complexity and

importance of such applications, it is a common practice to

perform an automatic (static) analysis of properties related

to anticipated application executions.

As a simple example, consider a typical E-commerce ap-

plication where users can navigate through a selection of

∗

This research was partially supported by the Frankel In-

stitute, the National Science Foundation (NSF IIS 1217798)

and the US-Israel Binational Science Foundation

Permission to make digital or hard copies of all or part of this work for

personal or classroom use is granted without fee provided that copies are

not made or distributed for proﬁt or commercial advantage and that copies

bear this notice and the full citation on the ﬁrst page. To copy otherwise, to

republish, to post on servers or to redistribute to lists, requires prior speciﬁc

permission and/or a fee. Articles from this volume were invited to present

their results at The 39th International Conference on Very Large Data Bases,

August 26th - 30th 2013, Riva del Garda, Trento, Italy.

Proceedings of the VLDB Endowment, Vol. 6, No. 12

products (classiﬁed through categories and sub-categories),

and proposed discount deals. The underlying process logic

can be captured via a Finite State Machine whose states

reﬂect web-pages and logical application states, and (part

of) its transitioning is governed by queries on an underlying

relational database. Now, an analyst may be interested in

answering, among others, the following questions:

• (A1) What is the probability that a user would exit

without viewing the daily discount deals?

• (A2) What is the minimal number of clicks allowing a

user to view daily deals?

• (A3) What is the probability that a user would view

product sub-categories for some category?

Such analysis tasks can be expressed in LTL (Linear Tem-

poral Logic) [12] which is commonly used in the context of

static analysis. To obtain a better understanding of the ap-

plication, analysts may want to do more than pose a single,

static analysis task with respect to its current speciﬁcation,

but rather to test and explore the eﬀect of diﬀerent modiﬁca-

tions to the process or underlying data, on analysis results.

Hence, she may want to perform analysis of the above ﬂavor

under hypothetical scenarios such as:

• (S1) What if we remove/add a link between two web

pages?

• (S2) What if a particular product is no longer avail-

able?

• (S3) What if the user is a member of a particular club?

In realistic cases, there may be many possible scenarios

and combinations thereof that are of interest. An analyst

would like to interactively explore the diﬀerent combina-

tions, reﬁning them according to the analysis results (e.g.

removing one link at a time and observing is eﬀect). A sim-

ple way to support the exploration of such scenarios is to ap-

ply each requested hypothetical scenario to the application

speciﬁcation, and then analyze the obtained speciﬁcation.

This, however, would be prohibitively ineﬃcient, as it in-

volves repeated costly evaluation of the LTL formula as well

as costly accesses to the application’s underlying database.

Instead, we propose an alternative approach and imple-

ment it in PROPOLIS (PROvisioned PrOcess anaLysIS), whose

high-level architecture is depicted in Fig. 1. Once the model

speciﬁcation was set, the analyst may (1) pose an LTL query

and (2) deﬁne hypothetical scenarios by deﬁning parameters

with respect to the process speciﬁcation (logical ﬂow and/or

underlying database) at points of interest. Then, PROPOLIS

evaluates the given LTL formula with respect to the param-

eterized speciﬁcation, and computes an expression referred

to as a provisioned expression. This expression captures, in

1302

下载后可阅读完整内容，剩余3页未读，立即下载

Leon12234567

粉丝: 0
资源: 7

PROPOLIS：数据驱动过程的假设分析系统

Laravel开发-lara-lambda .zip

LVM存储虚拟化_Thin-provisioned[参考].pdf

AWS -2022 SAA C02 .pdf

VMware VCP_2V0-21.20 V13.75.pdf

Device_Provisioned_issue_analysis_report

Performance Evaluation of Web-based Cloud Services in a Browser-Scripting Approach

VMware-vSphere日常操作手册.doc

AWS SAA-C02专家系统文档.docx

IBM Spectrum Virtualize HyperSwap-高可用解决方案.docx

最新资源