Question

I am writing a web application on google app engine in python using jinja2 as a templating system and the app basically allows users to write posts/comments and rank other user’s posts. The ranking system is based on the number of up/downvotes and the number of comments. I am trying to use memcache to store this calculated value and rank the posts accordingly.

I only want to store the value in the database occasionally so as to not make the write costs expensive. I am planning on having a counter and storing it in the database every 10 votes/comments. I was thinking something like this:

# I update the counter every time that I add a vote or comment
counter = 0
def posts_cache(update = False):
        global counter
        key = 'main'
        posts = memcache.get(key)
        if posts is None or update:
                logging.error("DB QUERY")
                posts = db.GqlQuery("SELECT * "
                                        "FROM Post "
                                        "ORDER BY rank DESC "
                                        "LIMIT 100",
                                         key)
                posts = list(posts)
                memcache.set(key, posts)
        if counter>=10:
                counter = 0
                #put the memcache posts in the database
        return posts

But I am not sure how to take the posts I have in memcache and store them in the database. Is there any way to do this in python? I’ve looked through the docs but haven’t found a clear way to do it.

Answer 1

1. You can’t do it because memcache is not a reliable in the sense it can evict your entities and than you will lost data.
2. Storing the entities in bulk don’t cost you less, you pay per entity put.
3. In order to lower your writing cost try setting indexed=false on attributes that don’t you don’t need them indexed.