Challenges and complexities in application of LCA approaches in the case of ICT for a sustainable future
1 like621 views
The document discusses the complexities of applying Life Cycle Assessment (LCA) approaches within the context of Information and Communication Technology (ICT) for sustainable development. It emphasizes a holistic model incorporating multiple actors and perspectives, advocating for a shift from a pillar-based framework to one that includes individual and enterprise contributions to sustainability. The text outlines the need for new metrics to assess the impact of technologies, particularly in relation to their environmental footprints and contextual sensitivities.
Challenges and complexities in application of LCA approaches in the case of ICT for a sustainable future
- 1. Challenges and complexities in application of LCA approaches in the case of ICT for a sustainable future Reza Farrahi Moghaddam, Fereydoun Farrahi Moghaddam, Thomas Dandres, Yves Lemieux, Réjean Samson, Mohamed Cheriet ICT4S’14, August 24-27, 2014, Stockholm, Sweden August 24, ConverStation I
- 2. Sustainability Pentagon • 3-pillar sustainability model: • ‘Respect’ to and ‘focus’ on all three aspects: Nature, Society, and Economy • Extended sustainability model: • To include all actors and actions: Actor/Action Model • Moving from ‘pillars’ to ‘actors’: society societies • Adding two additional categories for smaller actors: • Enterprises (businesses) and individuals (people) • Multiple ‘perspectives’ could be drawn from it • ‘Inclusion’ and ‘Engagement’ by ‘Understanding’ 2
- 3. Sustainability Pentagon, Cont’d • More categories can be added in future • Many actors, such as governments, would not require new categories, and will be mapped using ‘decomposability’ approach • Everybody 1. Inclusion: Everybody-Together 2. Responsibility: Everybody-in-the-Loop • Moving beyond reports and projects • Sustainable ‘way’ of business and living • Seamless integration of LCA and other approaches 3
- 4. Differential Metric/distance for Complex Systems • Enormous number of components • Many components follow the same ‘internal’ model • Common practice: Aggregating their footprint • However, they may be in different ‘contexts’ • We represent the context with some ‘attributes’ assigned to a component • Modeling the context (neighborhood) would be costly, therefore ‘over simplified’ practices are common • However, context changes are: ‘nonlinear’, large-scale, and complex • The need for a new distance function/metric to ‘preserve’ the impact of those components with small self ecological footprint (EF) but with big contextual sensitivity and therefore potential EF 4
- 5. Differential Metric/distance for Complex Systems, Cont’d • Assessment of a Technology vs Assessment of a Product: • An application of the proposed metric • The upstream or specific region could be ‘masked’ using the ‘attributes’ • The attributes of components are different from the attributers of footprint: • Attributes of footprint: Energy Consumption, GHG Emissions, etc. • Attributes of components: Geo Location, Systemic Location, Time, etc. • Similar debate on electric vehicle (EV) can be analyzed using the proposed approach: • The purpose is not to give the green light to a technology: • It is to see how that technology could approach to ‘zero’ impact • ‘Zero footprint’ is the new ‘footprint reduction’ 5
- 6. Differential Metric/distance for Complex Systems, Cont’d G1: Baseline graph (G2, G1): Differential graph G1 + attributes (not mergeable) The new distance function: • The full paper will be ready by the next week. • Application: Micro- registration and multicasting in OTT SVoD/IPTV 6
- 7. ICT Specialty: Management/Intelligence (M/I) • ICT: A perfect example of businesses based on ‘intangible’ properties • The physics of intangible properties: There is ‘zero’ physics • Considerable footprint assoc. with: • processing of properties’ information • Great potential for footprint reduction: • Smart Management even without hardware upgrade • However, considerable investment to modify low-level software • Specific Example: ‘Smart DVFS’, a step beyond DVFS and virtualization 7
- 8. ICT Specialty: Management/Intelligence (M/I), Cont’d: Smart DVFS • DVFS has been around: • An alternative to consolidation • Higher availability/Lower reaction time • Less footprint: i) less electricity consumption (EC), ii) less hardware replacement, and iii) less network and storage traffic and therefore less associated EC for VM image transfers. • Beyond VMs: Containers (Virtualization of ‘processes’) • Next step: Smart DVFS • Multi-objective optimization w.r.t. EC, profit, and EF • Smart heuristic scheduling of tasks (jobs) • Electricity Grid: mix and price vary over time (hours) and (zones) • A network of geographically-distributed data centers 8
- 9. ICT Specialty: Management/Intelligence (M/I), Cont’d: Smart DVFS • All methods are DVFS-enabled • CPAS: Proposed smart DVFS (Carbon-Profit Aware Scheduler) 9
- 10. Fast Links to Papers and Web Pages This Paper (http://arxiv.org/abs/1403.2798) MEIO Paper (http://arxiv.org/abs/1404.6227) Profile (http://ca.linkedin.com/in/rezafm) DVFS Thesis (http://espace.etsmtl.ca/1274/1/FARRAHI_MOGH ADDAM_Fereydoun.pdf) QoE – Multi-Tenant Resource Sharing (http://arxiv.org/abs/1407.5527v3) New-distance Paper TBF 10