Site Reliability Engineers (SREs) play a key role in issue identification and resolution. After an issue is reported, SREs come together in a virtual room (collaboration platform) to triage the issue. While doing so, they leave behind a wealth of inf
ormation which can be used later for triaging similar issues. However, usability of the conversations offer challenges due to them being i) noisy and ii) unlabelled. This paper presents a novel approach for issue artefact extraction from the noisy conversations with minimal labelled data. We propose a combination of unsupervised and supervised model with minimum human intervention that leverages domain knowledge to predict artefacts for a small amount of conversation data and use that for fine-tuning an already pretrained language model for artefact prediction on a large amount of conversation data. Experimental results on our dataset show that the proposed ensemble of unsupervised and supervised model is better than using either one of them individually.
Conversational channels are changing the landscape of hybrid cloud service management. These channels are becoming important avenues for Site Reliability Engineers (SREs) %Subject Matter Experts (SME) to collaboratively work together to resolve an in
cident or issue. Identifying segmented conversations and extracting key insights or artefacts from them can help engineers to improve the efficiency of the incident remediation process by using information retrieval mechanisms for similar incidents. However, it has been empirically observed that due to the semi-formal behavior of such conversations (human language) they are very unique in nature and also contain lot of domain-specific terms. This makes it difficult to use the standard natural language processing frameworks directly, which are popularly used in standard NLP tasks. %It is important to identify the correct keywords and artefacts like symptoms, issue etc., present in the conversation chats. In this paper, we build a framework that taps into the conversational channels and uses various learning methods to (a) understand and extract key artefacts from conversations like diagnostic steps and resolution actions taken, and (b) present an approach to identify past conversations about similar issues. Experimental results on our dataset show the efficacy of our proposed method.
We have studied the origin of a counter intuitive diffusion behavior of Fe and N atoms in a iron mononitride (FeN) thin film. It was observed that in-spite of a larger atomic size, Fe tend to diffuse more rapidly than smaller N atoms. This only happe
ns in the N-rich region of Fe-N phase diagram, in the N-poor regions, N diffusion coefficient is orders of magnitude larger than Fe. Detailed self-diffusion measurements performed in FeN thin films reveal that the diffusion mechanism of Fe and N is different - Fe atoms diffuse through a complex process, which in addition to a volume diffusion, pre-dominantly controlled by a fast grain boundary diffusion. On the other hand N atoms diffuse through a classical volume-type diffusion process. Observed results have been explained in terms of stronger Fe-N (than Fe-Fe) bonds generally predicted theoretically for mononitride compositions of transition metals.
In this work, we studied phase formation, structural and magnetic properties of iron-nitride (Fe-N) thin films deposited using high power impulse magnetron sputtering (HiPIMS) and direct current magnetron sputtering (dc-MS). The nitrogen partial pres
sure during deposition was systematically varied both in HiPIMS and dc-MS. Resulting Fe-N films were characterized for their microstructure, magnetic properties and nitrogen concentration. We found that HiPIMS deposited Fe-N films show a globular nanocrystalline microstructure and improved soft magnetic properties. In addition, it was found that the nitrogen reactivity impedes in HiPIMS as compared to dc-MS. Obtained results can be understood in terms of distinct plasma properties of HiPIMS.
