Search for: All records

Expanding electricity access (Sustainable Development Goal (SDG) 7) and empowering women (SDG 5) are closely linked. Most studies quantifying the benefits of the former for women focus on their economic empowerment; however, if and how such access results in women's empowerment is best understood by examining the cultural context, norms, and gender roles in which that access occurs. For instance, time saved from the use of electric appliances may be used for productive engagements, but if gender roles restrict women from leaving the home or engaging in paid work, such benefits are not realized. Here, we delve deeper into the multi-faceted and context-specific concept of women's empowerment via 28 semi-structured interviews with Zambian women. We include households with and without electricity to understand women's subjective meaning of empowerment and how access to electricity may (dis) empower them. We analyze their responses using Deshmukh-Ranadive's (2005) Spaces approach to empowerment which categorizes an individual's spaces into physical, economic, political, socio-cultural, and mental space. We find that electricity access empowers women by expanding their economic and physical, along with mental, space. This occurs via paid opportunities outside the home using electrical appliances and women reporting greater economic independence, camaraderie, self-reliance, and agency as a result. Additionally, by asking women to define what empowerment means to them, we not only bolster the claim that electricity access empowers women both economically and socially, but also ensure future programs account for empowerment explicitly in their plans. 

Sepsis and trauma are known to disrupt gut bacterial microbiome communities, but the impacts and perturbations in the fungal (mycobiome) community after severe infection or injury, particularly in patients experiencing chronic critical illness (CCI), remain unstudied.

We assess persistence of the gut mycobiome perturbation (dysbiosis) in patients experiencing CCI following sepsis or trauma for up to two-to-three weeks after intensive care unit hospitalization.

We show that the dysbiotic mycobiome arrays shift toward a pathobiome state, which is more susceptible to infection, in CCI patients compared to age-matched healthy subjects. The fungal community in CCI patients is largely dominated byCandidaspp; while, the commensal fungal species are depleted. Additionally, these myco-pathobiome arrays correlate with alterations in micro-ecological niche involving specific gut bacteria and gut-blood metabolites.

The findings reveal the persistence of mycobiome dysbiosis in both sepsis and trauma settings, even up to two weeks post-sepsis and trauma, highlighting the need to assess and address the increased risk of fungal infections in CCI patients.

 

We present ContextPrefetcher, a host-guided high-performant prefetching framework for near-storage accelerators that prefetches data blocks from storage (e.g., NAND) to devicelevel RAM. Efficiently prefetching data blocks to device-level RAM reduces storage access costs and improves I/O performance. We introduce a novel abstraction, Cross-layered Context (CLC), a virtual entity that spans across the host and the device and is used for identifying, managing, and tracking active and inactive data such as files, objects (within object stores), or a range of blocks. To support efficient prefetching of actively used CLCs to device memory without incurring near-device resource (memory and compute) bottlenecks, ContextPrefetcher delegates prefetching management to the host, guiding near-device compute to prefetch blocks of active CLC. Finally, ContextPrefetcher facilitates the swift reclamation of blocks associated with inactive CLC. Preliminary evaluation against state-of-the-art near-storage accelerator designs demonstrates performance gains of up to 1.34×. 

Graph analytics shows promise for solving challenging problems on relational data. However, memory constraints arise from the large size of graphs and the high complexity of algorithms. Data prefetching is a crucial technique to hide memory access latency by predicting and fetching data into the memory cache beforehand. Traditional prefetchers struggle with fixed rules in adapting to complex memory access patterns in graph analytics. Machine learning (ML) algorithms, particularly long short-term memory (LSTM) models, excel in memory access prediction. However, they encounter challenges such as difficulty in learning interleaved access patterns and high storage costs when predicting in large memory address space. In addition, there remains a gap between designing a high-performance ML-based memory access predictor and developing an effective ML-based prefetcher for an existing memory system. In this work, we propose a novel Attention-based prefetching framework to accelerate graph analytics applications. To achieve high-performance memory access prediction, we propose A2P, a novel Attention-based memory Access Predictor for graph analytics. We use the multi-head self-attention mechanism to extract features from memory traces. We design a novelbitmap labelingmethod to collect future deltas within a spatial range, making interleaved patterns easier to learn. We introduce a novelsuper pageconcept, allowing the model to surpass physical page constraints. To integrate A2P into a memory system, we design a three-module prefetching framework composed of an existing memory hierarchy, a prefetch controller, and the predictor A2P. In addition, we propose a hybrid design to combine A2P and existing hardware prefetchers for higher prefetching performance. We evaluate A2P and the prefetching framework using the widely used GAP benchmark. Prediction experiments show that for the top three predictions, A2P outperforms the widely used state-of-the-art LSTM-based model by 23.1% w.r.t. Precision, 21.2% w.r.t. Recall, and 10.4% w.r.t. Coverage. Prefetching experiments show that A2P provides 18.4% IPC Improvement on average, outperforming state-of-the-art prefetchers BO by 17.2%, ISB by 15.0%, and Delta-LSTM by 10.9%. The hybrid prefetcher combining A2P and ISB achieves 21.7% IPC Improvement, outperforming the hybrid of BO and ISB by 16.3%.