No description available.

La tendencia general de los computadores paralelos es crecer en complejidad y en número de componentes. La miniaturización y la concentración de dichos elementos es la principal causa de la aparición y aumento de los fallos en estos computadores. Asimismo, para permitir la ejecución correcta de las aplicaciones paralelas, existe la necesidad de proveer soporte y de tolerar fallos en estos entornos. Una estrategia amplamente utilizada es el rollback-recovery, que consiste en guardar periódicamente el estado de la aplicación y, en caso de fallos, reanudar la aplicación desde el último estado guardado. El uso de estos protocolos añade una sobrecarga al tiempo de ejecución de la aplicación. Con el uso de protocolos de checkpoints no coordinados, es fácil estimar el tiempo total de ejecución de una aplicación, así como también la frecuencia en la cual estos checkpoints deben ser guardados. Actualmente, existen modelos precisos para estimar estos tiempos. Sin embargo, el uso de protocolos de checkpoints coordinados, puede no ser la mejor solución para proveer tolerancia a fallos en los computadores paralelos de próxima generación. En otras palabras, el actual paradigma de tolerancia a fallos para computadores paralelos, no es adecuado para los futuros sistemas. Los protocolos de tolerancia a fallos no coordinados permiten que, cada proceso de la aplicación paralela guarde su estado independientemente de los demás procesos; la combinación de estos protocolos con técnicas de log de eventos eliminan los inconvenientes de los protocolos no coordinados, como el efecto domino y la aparición de mensajes huérfanos. Esta combinación representa el paradigma emergente de tolerancia a fallos para aplicaciones paralelas escalables. Actualmente, no hay modelos adecuados para estimar el tiempo de ejecución de aplicaciones paralelas que están siendo protegidas por checkpoints no coordinados. Así como tampoco existen modelos para calcular la frecuencia en que dichos checkpoints deben ser creados. El objetivo de esta tesis es, definir los modelos específicos para cada uno de los paradigmas: el coordinado y el no coordinado. Los modelos proveen una estimación del tiempo total de ejecución de las aplicaciones cuando están protegidas por cualquiera de los dos paradigmas. Además, se propone una metodología para definir el valor de las variables necesarias para calcular el intervalo de checkpoints. La principal motivación de este trabajo es proveer el conocimiento necesario para enfrentar el paradigma emergente de tolerancia a fallos y hacerlo asequible para los usuarios de las aplicaciones paralelas. / Parallel computers are growing in complexity and in number of components. The components miniaturisation and concentration are the major root causes of the failures increasingly seen on these computers. Thus, in order to achieve the execution end, parallel application should use a fault tolerance strategy. A widely used strategy is the rollback-recovery, which consists of saving the application state periodically. In the event of a fault occurring, the application resumes it execution from the most recent saved state. These fault tolerance protocols include an overhead on the parallel application execution. Using a coordinated checkpointing protocol it becomes easy to estimate the application execution time, as well as to calculate the frequency in which checkpoints should be taken. In fact, there are very precise models to estimate the application execution time and the checkpoint interval nowadays. However, the use of the coordinated checkpointing may not be the best solution to provide fault tolerance on the next-generation parallel computers. In other words, the current paradigm of fault tolerance for parallel applications is not suitable for the future parallel computer. Fault tolerance protocols such as uncoordinated checkpointing permits that each process of the parallel application saves its state independently of other processes. The combination of uncoordinated checkpointing with logging of message-passing events avoids the inconvenience of this sort of protocol, such as the domino effect and orphan messages. This is the emergent paradigm of fault tolerance for scalable parallel applications. For instance, there is no model suitable to estimate the execution time of a parallel application protected by uncoordinated checkpointing. As well as there is no convenient model to calculate the frequency in which those checkpoints should be taken. The objective of this thesis is to define suitable models that can be used with each paradigm: the coordinated and the uncoordinated. These models should provide an estimation of the application wall time clock running under each fault tolerance paradigm, as well a methodology to define the value of the variables used to calculate the checkpointing interval. The main motivation of this work is to provide at the same time the knowledge necessary to face the emergent fault tolerance paradigm and make it suitable to be used by parallel applications users.

The conventional computing techniques are based on the assumption of a near constant source of input power. While this assumption is reasonable for high-end devices such as servers and mobile phones, it does not always hold in embedded devices. An increasing number of Internet of Things (IoTs) is powered by intermittent power supplies which harvest energy from ambient resources, such as vibrations. While the energy harvesters provide energy autonomy, they introduce uncertainty in input power. Intermittent computing techniques were proposed as a coping mechanism to ensure forward progress even with frequent power loss. They utilize non-volatile memory to store a snapshot of the system state as a checkpoint. The conventional security mechanisms do not always hold in intermittent computing. This research takes a top-down approach to design secure intermittent systems. To that end, we identify security threats, design a secure intermittent system, optimize its performance, and evaluate our design using embedded benchmarks. First, we identify vulnerabilities that arise from checkpoints and demonstrates potential attacks that exploit the same. Then, we identify the minimum security requirements for protecting intermittent computing and propose a generic protocol to satisfy the same. We then propose different security levels to configure checkpoint security based on application needs. We realize configurable intermittent security to optimize our generic secure intermittent computing protocol to reduce the overhead of introducing security to intermittent computing. Finally, we study the role of application in intermittent computing and study the various factors that affect the forward progress of applications in secure intermittent systems. This research highlights that power loss is a threat vector even in embedded devices, establishes the foundation for security in intermittent computing. / Doctor of Philosophy / The embedded systems are present in every aspect of life. They are available in watches, mobile phones, tablets, servers, health aids, home security, and other everyday useful technology. To meet the demand for powering up a rising number of embedded devices, energy harvesters emerged as a solution to provide an autonomous solution to power on low-power devices. With energy autonomy, came energy scarcity that introduced intermittent computing, where embedded systems operate intermittently because of lack of constant input power. The intermittent systems store snapshots of their progress as checkpoints in non-volatile memory and restore the checkpoints to resume progress. On the whole, the intermittent system is an emerging area of research that is being deployed in critical locations such as bridge health monitoring. This research is focused on securing intermittent systems comprehensively. We perform a top-down analysis to identify threats, mitigate them, optimize the mitigation techniques, and evaluate the implementation to arrive at secure intermittent systems. We identify security vulnerabilities that arise from checkpoints to demonstrate the weakness in intermittent systems. To mitigate the identified vulnerabilities, we propose secure intermittent solutions to protect intermittent systems using a generic protocol. Based on the implementation of the generic protocol and its performance, we propose several optimizations based on the needs of the application to securing intermittent systems. And finally, we benchmark the security properties using two-way relation between security and application in intermittent systems. With this research, we create a foundation for designing secure intermittent systems.

E4F1 est un facteur de transcription liant l'ADN, exprimé de façon ubiquitaire par tous les tissus, et qui possède une activité E3 ubiquitine ligase atypique dirigée contre le suppresseur de tumeur p53. La protéine E4F1 interagit directement avec plusieurs suppresseurs de tumeurs cellulaires et des oncogènes viraux (p53, pRb, DRAL, RASSF1A, p19ARF, BMI1, HBX et GAM1...), suggérant qu’elle est elle-même impliquée dans la tumorigenèse. La perte d’E4F1 dans des fibroblastes embryonnaires (Mefs) transformés ou dans des cellules de sarcomes histiocytaires déficientes pour la voie p53, entraine la mort de ces cellules. La même inactivation d'E4F1 dans les cellules normales n'affecte pas leur survie mais entraine un arrêt de prolifération. Des analyses transcriptomiques et de liaison à l'ADN à l'échelle du génome entier (ChIP-seq et analyses différentielles des transcriptomes de cellules E4F1 WT et KO) nous ont permis d’identifier une centaine de gènes liés et régulés directement par E4F1. Ces gènes codent notamment pour des protéines mitochondriales impliquées dans le métabolisme et l'homéostasie de cette organelle, dont plusieurs composants et régulateurs de l'enzyme multimérique, pyruvate déshydrogénase. Un second groupe de gênes cibles d’E4F1 est impliqué dans la réponse aux dommages à l'ADN, dont le gène codant pour la kinase CHK1 qui joue un rôle essentiel dans le contrôle de la stabilité du génome. En accord avec la fonction de ces gènes cibles, la perte d’E4F1 entraine des perturbations du métabolisme cellulaire et des checkpoints de réponse aux stress génotoxiques. Dans les cellules déficientes pour la voie p53, ces perturbations conduisent à des stress oxydatifs (surproduction de ROS mitochondriaux) et énergétiques, suivis de dommages aux protéines et à l'ADN, et in fine, à une mort cellulaire massive. Dans les cellules compétentes pour la voie p53 ces altérations sont fortement atténuées et conduisent à un arrêt de la prolifération. Une partie des effets protecteurs de p53 est due à sa capacité à stimuler l'expression du gène ALDH4a1 qui code pour une aldehyde dehydrogenase impliquée dans le catabolisme de la proline et dont l'activité possède des propriétés anti-oxydantes.Mes travaux mettent également en évidence que le niveau d'expression de la protéine E4F1 et son activité augmentent lors de la transformation cellulaire ainsi qu'en réponse à des stress énergétiques, génotoxiques ou oxydatifs. Dans ces trois dernières conditions, la phosphorylation d'E4F1 est également augmentée au niveau de plusieurs sérines qui ont été identifiées par spectrométrie de masse. En résumé, tous ces éléments indiquent qu'E4F1 est un acteur important du contrôle de l'homéostasie métabolique et de la réponse aux stress, particulièrement essentiel pour la survie des cellules cancéreuses déficientes pour la voie p53. Mes observations suggèrent également qu'E4F1 est activé en réponse à différents stress et qu'il pourrait jouer un rôle essentiel dans la capacité des cellules cancéreuses à s'adapter aux multiples stress environnementaux auxquels elles sont exposées au cours de la tumorigenèse. / The ubiquitously expressed E4F1 protein acts as a transcription factor that binds a consensus DNA sequence at promoters, and as an atypical E3-ligase for the tumor suppressor p53. E4F1 physically interacts with several bona fide cellular tumor suppressors and viral oncoproteins (including p53, pRb, DRAL, RASSF1A, p19ARF, BMI1, HBX and GAM1...), suggesting that it might itself be involved in tumorigenesis. E4F1 genetic inactivation in transformed mouse embryo fibroblasts (Mefs) and in hematopoietic tumors deficient for the p53 pathway, results in massive cell death. Importantly, inactivation of E4F1 in normal cells does not affect cell survival. Genome wide approaches (ChIP-seq profiling and comparative transcriptomics performed on E4F1 WT and KO cells) identified a limited list (100) of genes that are bound and directly regulated by E4F1. Several E4F1 target genes code for mitochondrial proteins involved in mitochondria metabolism and homeostasis, including several components of the pyruvate dehydrogenase complex. Another set of E4F1 target genes codes for factors involved in DNA repair and damage checkpoints, including the checkpoint kinase CHK1. Accordingly, both mitochondrial and checkpoint functions are altered in E4F1 KO cells. In proliferating cells deficient for the p53 pathway, these defects lead to energetic and oxidative stresses, protein and DNA damages, and in fine, massive cell death. In p53-proficient cells, these alterations are attenuated and lead to growth arrest. Part of this protective effect of p53 is mediated by ALDH4a1, a p53 target gene encoding an aldehyde dehydrogenase involved in proline catabolism and that exhibits antioxidant properties.In this thesis, I also demonstrate that E4F1 protein level and activity is increased during cell transformation and upon exposure to genotoxic, energetic or oxidative stresses. These stresses also lead to E4F1 phosphorylation at specific serine residues that were identified by mass spectrometry. All together this work shows that E4F1 controls cellular functions that are important for mammalian cells metabolic homeostasis and stress responses, and that are essential for the survival of p53-deficient cancer cells. This work also suggests that E4F1 is activated in response to various stresses and therefore, that it could play an essential role in allowing cancer cells to adapt to environmental stresses.

Le cancer colorectal est responsable de plus de 17500 décès par an et se situe à la 3ème place des cancers les plus fréquents en France. En altérant le processus de réplication de l’ADN des cellules cancéreuses, certaines des molécules utilisées en chimiothérapie induisent du stress réplicatif. Au niveau cellulaire, ce stress est géré par la voie ATR-CHK1. Quand elle est activée par des régions d’ADN simple brin protégées par RPA au niveau de fourches de réplication ralenties/bloquées, ATR déclenche l’activation du checkpoint intra-S via la phosphorylation de son effecteur CHK1. Ce checkpoint permet alors aux cellules de gérer ce stress réplicatif, via différents processus (arrêt du cycle cellulaire, régulation de l’allumage d’origines de réplication, stabilisation des fourches…). Depuis quelques années, l’intérêt thérapeutique de cibler cette voie est clairement établi dans la littérature. Ce rationnel thérapeutique repose sur une inhibition pharmacologique de la voie ATR-CHK1, éventuellement couplée à des traitements par des molécules génotoxiques. Par ailleurs, certaines tumeurs présentent fréquemment des mutations hétérozygotes des gènes ATR et CHK1. Nous avons émis l’hypothèse que ces déficiences puissent représenter leur talon d’Achille. L’équipe a généré un modèle cellulaire permettant d’étudier spécifiquement ces mutations hétérozygotes d’ATR et CHK1. Nous avons commencé notre étude par la caractérisation des impacts de ces mutations en conditions normales de culture. Nos données montrent que les mutations d’ATR et CHK1 altèrent l’activation basale du checkpoint intra-S, provoquent un stress réplicatif endogène, et aboutissent à une induction de dommages de l’ADN. Par ailleurs, ces mutations sensibilisent les cellules à certaines drogues. Entre autres, les cellules mutantes présentent des sensibilités cytotoxiques accrues au SN-38 (principe actif de l’Irinotécan, inhibiteur de topoisomérase 1) et au VE-822 (inhibiteur d’ATR). De plus, nous avons montré que ces deux composés ont un effet synergique important, et nous avons par la suite étudié les mécanismes moléculaires sous-jacents à ces phénotypes de sensibilisation et de synergie. Nos résultats démontrent que la combinaison SN-38+VE-822 entraîne une apoptose dépendante de la caspase-3, exacerbée chez les cellules mutantes ATR ou CHK1. Ces altérations génétiques limitent le potentiel d’activation du checkpoint intra-S et aboutissent à une accumulation de dommages de l’ADN. L’inhibition d’ATR par le VE-822 permet aux cellules de court-circuiter l’arrêt du cycle cellulaire en S-précoce normalement induit par le SN-38. Nos analyses démontrent que ce phénotype entraine un épuisement de RPA et une catastrophe réplicative subséquente, la mutation d’ATR prédisposant les cellules à ces phénotypes. Les cellules survivant à la combinaison SN-38+VE-822 complètent la réplication et s’accumulent en G2 de façon DNA-PK-dépendante. Ce checkpoint post-réplicatif protège les cellules de la catastrophe mitotique. Ensemble, ces observations suggèrent que RPA et DNA-PK représentent des cibles thérapeutiques prometteuses pour optimiser les effets de l’inhibition de la voie ATR-CHK1. En définitive, les mutations d’ATR et CHK1 retrouvées chez les patients pourraient représenter des facteurs pronostiques importants de la réponse à ces stratégies thérapeutiques. De plus, certains de nos résultats suggèrent également une implication de la voie ATR-CHK1 dans la régulation du remodelage des fourches de réplication, notamment dans la résection de l’ADN néo-synthétisé. En affinant la compréhension des processus moléculaires impliqués dans la réponse au stress réplicatif, notre étude pourrait contribuer à l’amélioration de la prise en charge thérapeutique du cancer colorectal. / Colorectal cancer is responsible for more than 17,500 deaths per year and ranks third among the most frequent cancers in France. By interfering with the DNA replication process of cancer cells, several chemotherapeutic molecules induce replication stress. At the cellular level, this stress is managed by the ATR-CHK1 pathway. When activated by RPA-protected single-stranded DNA regions at slowed/blocked replication forks, ATR triggers the activation of the intra-S checkpoint via the phosphorylation of its CHK1 effector. This checkpoint then allows the cells to manage this replicative stress, via different processes (stopping the cell cycle, regulating the ignition of replication origins, stabilizing the forks...). In recent years, the therapeutic value of targeting this pathway has been clearly established in the literature. This therapeutic rationale is founded on pharmacological inhibition of the ATR-CHK1 pathway, possibly coupled with genotoxic molecules treatments. In addition, some tumours frequently have heterozygous mutations of the ATR and CHK1 genes. We have hypothesized that these deficiencies may represent their Achilles' heel. Our team generated a cellular model to specifically study these heterozygous mutations of ATR and CHK1. We began our study by characterizing the impacts of these mutations under normal growing conditions. Our data show that the ATR and CHK1 mutations alter the basal activity of the intra-S checkpoint, cause endogenous replicative stress, and lead to spontaneous DNA damage. In addition, these mutations sensitize the cells to certain drugs. Amongst other things, mutant cells show increased cytotoxic sensitivities to SN-38 (active ingredient of Irinotecan, topoisomerase inhibitor 1) and to VE-822 (ATR inhibitor). Furthermore, we showed that these two compounds have a strong synergistic. We then studied the underlying molecular mechanisms to these sensitization and synergy phenotypes. Our results show that the combination SN-38+VE-822 causes caspase-3-dependent apoptosis, exacerbated in mutant ATR or CHK1 cells. These genetic alterations limit the activation potential of the intra-S checkpoint and lead to extensive DNA damages. Inhibition of ATR by VE-822 allows cells to bypass the S- early cell cycle arrest normally induced by SN-38. Our analyses show that this phenotype leads to RPA depletion and subsequent replicative catastrophe, with ATR mutation predisposing cells to these phenotypes. Cells surviving the SN-38+VE-822 combination complete the replication and accumulate to G2 in a DNA-PK-dependent manner. This post-replicative checkpoint protects the cells from mitotic catastrophe. Together, these data suggest that RPA and DNA-PK represent promising therapeutic targets to optimize the effects of inhibition of the ATR-CHK1 pathway. Moreover, some of our results also suggest that the ATR-CHK1 pathway could be involved in the regulation of replication forks' remodeling, particularly in the resection of newly-synthetized DNA. Ultimately, the mutations of ATR and CHK1 found in patients may represent important prognostic factors in the response to these therapeutic strategies. By achieving a better understanding of the molecular processes involved in the response to chemically-induced replication stress, our study could contribute to the improvement of colorectal cancer’s therapeutic management.

In this work, a new roll-forward check pointing scheme is proposed using basic checkpoints. The direct-dependency concept used in the communication-induced check pointing scheme has been applied to basic checkpoints to design a simple algorithm to find a consistent global checkpoint. Both blocking (i.e. when the application processes are suspended during the execution of the algorithm) and non-blocking approaches have been presented. The use of the concept of forced checkpoints ensures a small re-execution time after recovery from a failure. The proposed approaches enjoy the main advantages of both the synchronous and the asynchronous approaches, i.e. simple recovery and simple way to create checkpoints. Besides, in the proposed blocking approach direct-dependency concept is implemented without piggybacking any extra information with the application message. A very simple scheme for avoiding the creation of useless checkpoints has also been proposed.

The soil-nematode Caenorhabditis elegans survives oxygen deprivation (anoxia < 0.001 kPa of O2, 0% O2) by entering into a state of suspended animation during which cell cycle progression at interphase, prophase and metaphase stage of mitosis is arrested. I conducted cell biological characterization of embryos exposed to various anoxia exposure times, to demonstrate the requirement and functional role of spindle checkpoint gene san-1 during brief anoxia exposure. I conducted a synthetic lethal screen, which has identified genetic interactions between san-1, other spindle checkpoint genes, and the kinetochore gene hcp-1. Furthermore, I investigated the genetic and cellular mechanisms involved in anoxia-induced prophase arrest, a hallmark of which includes chromosomes docked at the nuclear membrane. First, I conducted in vivo analysis of embryos carried inside the uterus of an adult and exposed to anoxic conditions. These studies demonstrated that anoxia exposure prevents nuclear envelope breakdown (NEBD) in prophase blastomeres. Second, I exposed C. elegans embryos to other conditions of mitotic stress such as microtubule depolymerizing agent nocodazole and mitochondrial inhibitor sodium azide. Results demonstrate that NEBD and chromosome docking are independent of microtubule function. Additionally, unlike anoxia, exposure to sodium azide causes chromosome docking in prophase blastomeres but severely affects embryonic viability. Finally, to identify the genetic mechanism(s) of anoxia-induced prophase arrest, I conducted extensive RNA interference (RNAi) screen of a subset of kinetochore and inner nuclear membrane genes. RNAi analysis has identified the novel role of 2 nucleoporins in anoxia-induced prophase arrest.

Immune exhaustion can be a major clinical problem for patients who have cancer or chronic inflammation. Persistent antigen stimulation drives T cells to express multiple surface markers called immune checkpoints. When these markers bind to their corresponding ligands that are expressed by antigen (e.g. tumor cells), T cells become metabolically impaired and lose several important functions; some cell signaling pathways are inhibited, while other intracellular mediators are re-modulated. Eventually, both CD4+ and CD8+ T cells behave dysfunctionally in ways that may facilitate cancer progression. Immune checkpoints are a major hallmark of immune exhaustion. In addition, natural killer (NK) cells, a critical immune cell subset in the peripheral immune system, also express immune checkpoint molecules, and are responsible for detecting and destroying circulating tumor cells. Yet, little research has investigated immune checkpoints on NK cells. Here, we explored the role of Bromodomain and ExtraTerminal domain (BET) proteins (BRD2, BRD3, BRD4), which are important transcriptional co-regulators, and critical for proliferation and metastasis in many cancer types, in the regulation of immune checkpoint molecules in several immune cell subsets, including CD4+ and CD8+ T cells, and NK cells. Through binding to acetylated histone tails of nucleosomal chromatin, BET proteins assist in transcription of multiple genes. Deregulated expression of BET proteins promotes cancer development or tumor cell metastasis, and new data show the BET proteins contribute to immune exhaustion. Furthermore, Type 2 diabetes mellitus (T2DM) is another worrisome problem related to cancer. T2DM patients show increased risk of developing cancer. Patients with both T2DM and any type of cancer show higher risks for metastasis. Significantly, T2DM patients also show immune exhaustion, suggesting a hypothesis that BET proteins may couple immune system dysfunction, abnormal metabolism and cancer incidence or progression. Specifically, T2DM has been defined to be a metabolic and a chronic inflammatory disease. The 5' Adenosine Monophosphate-activated Protein Kinase (AMPK) signaling pathway is a key pivot of cell metabolism and as well a significant target of drugs that normalize blood glucose, such as metformin. Based on published data, we considered that it is important to explore the mechanism of how immune checkpoints are regulated through metabolic pathways, focusing on immune exhaustion in T2DM patients. Moreover, considering that the expression of BET proteins promotes cancer development and progression, and metastasis and immune exhaustion are characteristic of many cancers as well, we suspected a potential relationship among BET proteins, the AMPK metabolic signaling pathway and immune exhaustion is worth exploring. Here, we measure expression of the immune checkpoint molecules TIM-3, TIGIT, PD-1, and CTLA-4 on normal T cells and NK cells by flow cytometry. We demonstrate different degrees of regulation of immune checkpoints by BET proteins on stimulated T cells and NK cells. Comparing stimulated-only cells with stimulated-plus AMPK inhibitor cells, we found that inhibition of the AMPK signaling pathway causes divergent expression patterns for TIM-3 and TIGIT, PD-1 and CTLA-4. Simultaneous inhibition of both BET proteins and the AMPK signaling pathway, shows that BET proteins regulate TIM-3 and TIGIT through an AMPK-independent metabolic pathway and regulate PD-1 and CTLA-4 through an AMPK-dependent pathway. Overall, we show TIM-3 and TIGIT, PD-1 and CTLA-4 display different expression patterns under regulation of the AMPK signaling pathway, and we show that BET proteins regulate TIM-3, TIGIT, PD-1 and CTLA-4 through both AMPK-dependent and -independent pathways. These findings are important because they reveal novel mechanisms of immune checkpoint regulation, which may be valuable for targeting in cancer patients who are being treated with checkpoint inhibitors.

A transaction-consistent global checkpoint of a database records a state of the database which reflects the effect of only completed transactions and not the re- sults of any partially executed transactions. This thesis establishes the necessary and sufficient conditions for a checkpoint of a data item (or the checkpoints of a set of data items) to be part of a transaction-consistent global checkpoint of the database. This result would be useful for constructing transaction-consistent global checkpoints incrementally from the checkpoints of each individual data item of a database. By applying this condition, we can start from any useful checkpoint of any data item and then incrementally add checkpoints of other data items until we get a transaction- consistent global checkpoint of the database. This result can also help in designing non-intrusive checkpointing protocols for database systems. Based on the intuition gained from the development of the necessary and sufficient conditions, we also de- veloped a non-intrusive low-overhead checkpointing protocol for distributed database systems. Checkpointing and rollback recovery are also established techniques for achiev- ing fault-tolerance in distributed systems. Communication-induced checkpointing algorithms allow processes involved in a distributed computation take checkpoints independently while at the same time force processes to take additional checkpoints to make each checkpoint to be part of a consistent global checkpoint. This thesis develops a low-overhead communication-induced checkpointing protocol and presents a performance evaluation of the protocol.

In this study we pose the question if you through finding patterns in implementation of checkpoints can find ways to tailor the experience of the game A Story About My Uncle to hardcore gamers. To find out the answer we had hardcore players playing a modified level from the game A Story About My Uncle that was used as stimulus material for the following focus group interview. The analysis of the gathered material resulted in four patterns that can be used when implementing checkpoints to tailor the experience of the game A Story About My Uncle to hardcore gamers. Elements of checkpoints that our study showed to be important when it comes to how the game is experienced are density, placement and clearness. / I den här studien ställde vi oss frågan om man genom att hitta mönster inom implementation av checkpoints kunde finna sätt att anpassa spelupplevelsen i spelet A Story About My Uncle till hardcore spelare. För att ta reda på det lät vi hardcorespelare spela en modifierad bana ur spelet A Story About My Uncle som användes som stimulusmaterial för en följande fokusgruppsintervju. Analysen av det insamlade materialet resulterade i fyra mönster som kan användas vid implementering av checkpoints för att anpassa spelupplevelsen i spelet A Story About My Uncle till hardcorespelare. Faktorer för checkpoints som i vår studie visat sig ha stor betydelse i hur spelet upplevs är densitet, placering och tydlighet.